Vassiliki Poulou1; Lamprini Seremeti1
How can students be effectively introduced to the basic principles of prompt engineering in the age of generative artificial intelligence (GenAI)? This study approaches the question through an experiential learning design, using the analogy of the Pythia as a pedagogical entry point. The educational intervention was implemented as part of workshops conducted during the Panhellenic Student Conference entitled "Bring Your Own Prophecy to the Pythia". It was designed as a two-phase interactive learning experience, combining audiovisual material, short presentations, and participatory digital tools (live polling, word clouds, quizzes, and open-ended responses via mobile devices). In the first phase, students, aged 8–18, explored two heterogeneous "answer-generating systems", namely, the Pythia and GenAI, by examining key dimensions such as the role of the question, the nature of the response, the need for interpretation, the presence of uncertainty, and the "black-box of intermediator between question and answer". This comparative process enabled students to recognize that, while the Pythia requires interpretation and GenAI requires validation, both systems fundamentally depend on the formulation of the question. In the second phase, the workshop shifted toward the use of GenAI for content generation, where prompt engineering was introduced as a key practice for optimizing outputs. Through guided experimentation, students recognized that more precise and well-structured questions produce more relevant responses, thereby reducing the need for extensive validation. Findings indicate a clear cognitive shift, that is, students moved from focusing on evaluating answers to reflecting on how questions must be formulated. This shift highlights prompt engineering as a core cognitive competence of contemporary digital literacy.
Natalia Kougia1; Lamprini Seremeti1
Educational decision-making is increasingly relying on the processing of large volumes of heterogeneous data generated by multiple actors within the educational ecosystem, including students, teachers, administrators, policymakers, families, and other stakeholders. Such data may take both explicit and tacit forms, depending on their source, as they may originate from learning management systems, classroom interaction, institutional records, teacher observations, and policy-oriented databases. To process, interpret, and use these data effectively, several data-driven fields have emerged, including Educational Data Analytics, Learning Analytics, Educational Data Mining, Academic Analytics, Big Data in Education, Predictive Analytics, and Data-Driven Decision-Making. The insights generated from these fields are mainly used to support personalized learning, monitor student progress, improve curriculum design, and evaluate educational interventions. Nevertheless, the terminology used remains conceptually ambiguous. The relevant terms are often used interchangeably, although they differ in scope, technical approach, and pedagogical objective. This lack of semantic clarity may lead to inconsistent educational design. In this paper, we aim to clarify the conceptual landscape of data analytics in education, by reviewing the main research areas, defining the key terms used, and identifying the meaningful relationships among them.
Chakabwata William1
The rapid integration of Artificial Intelligence (AI) into higher education has generated debate regarding the future role of educators in the design and facilitation of learning. Some commentators such as Ertugan & Kucukuncular (2025), Du & Kohnke, (2025) and Kostanek & Li, (2025) argue that AI-driven systems may diminish the agency of educators by assuming functions traditionally associated with pedagogy, assessment, and curriculum design. Others point to the potential of AI to enhance teaching and learning through personalised learning pathways, increased efficiency, and expanded access to educational resources. Drawing on post human theory, this presentation challenges binary conceptions that position humans and machines as competing entities. Post humanism views human and technological actors as interconnected and mutually constitutive, creating opportunities for new forms of knowledge production and learning. From this perspective, AI need not be regarded as a threat to educators but rather as a tool that can augment human judgement, creativity, and pedagogical decision-making. The presentation will examine how AI can be integrated into higher education in ways that preserve human agency while leveraging technological capabilities. It argues that concerns regarding the displacement of educators often rest on assumptions about what it means to be human and about the exclusive ownership of pedagogical expertise. Rather than perceiving the relationship between humans and AI in oppositional terms, the paper proposes a collaborative model in which educators and AI systems work in partnership to support teaching, learning, and knowledge creation. The presentation concludes by considering the implications of such a model for pedagogy, academic identity, and the future of higher education.
The Assisted Climb Model and meaningful metacognitive education in the AI-industrial age
Sam Clarke1; Dawn Taylor2
As generative artificial intelligence (GenAI) increasingly reshapes how learners access, produce, and evaluate knowledge, schools face growing pressure to move beyond tokenistic approaches to metacognition that often reduce reflective thinking to isolated strategies or generic "learning to learn" activities. While such approaches served an important purpose within earlier educational contexts, they may be insufficient for preparing students for GenAI-mediated environments in which cognition and knowledge production are increasingly distributed across human and artificial systems. This paper presents findings from an immersive one-day pilot intervention based on the Assisted Climb model, implemented with Year 9 students across two secondary schools. Building on the reconceptualisation of Nottingham's Learning Pit for the age of GenAI, the intervention introduced students to the concepts of the Map, Torch, and Compass to support metacognitive awareness, epistemic judgement, and strategic regulation of GenAI use. Through collaborative activities, guided GenAI interaction, and structured reflection, learners were encouraged to evaluate knowledge claims, recognise cognitive offloading, and reflect on how GenAI shaped their thinking. Using a mixed-methods exploratory design, the study examined changes in self-awareness, confidence calibration, and epistemic agency before and after the intervention. Early findings suggest that the Assisted Climb model provided students with a more meaningful and transferable understanding of metacognition than many previously experienced school-based approaches. The paper argues that metacognitive education must evolve beyond procedural strategies toward forms of epistemic agency better aligned with the cognitive demands of the emerging AI-industrial age.
Ethical and Pedagogical Aspects of Generative Artificial Intelligence Use in Secondary Education from Students’ Perspective
Ivan Z. Davidović1
This paper explores the ethical and pedagogical aspects of the use of generative artificial intelligence in secondary education from students’ perspective. Given the increasingly widespread use of tools such as ChatGPT in school settings, the paper examines how students perceive the boundary between learning and copying, as well as the impact of these tools on their understanding of subject matter, independent learning, and the development of critical thinking.
Particular attention is paid to ethical dilemmas, including questions of academic integrity, responsibility, and fairness in the educational process. The research was conducted through an anonymous survey among students at Zemun Grammar School, examining the frequency and ways in which AI tools are used, their impact on learning, and students’ attitudes towards the ethical acceptability of their use.
The findings provide insight into students’ perceptions of the role of generative artificial intelligence in education and point to the need for clearer pedagogical guidelines and the development of AI literacy in schools.
Spandana Prem Deshpande1; Simran Mathur1; Rikhil Nagpal1
The growing use of generative artificial intelligence in fashion design education is reshaping the way students learn, ideate, and create. AI-powered tools such as image generators, trend prediction systems, and digital design assistants are increasingly becoming part of the creative learning process. While these technologies offer speed, accessibility, and new forms of inspiration, they also raise important concerns about originality, independent thinking, and the future role of human creativity in design education. This paper examines the relationship between human creativity and generative AI within fashion design education. It explores how students interact with AI-assisted tools during concept development, visual exploration, and design execution, while also questioning whether excessive dependence on such technologies may affect creative confidence, critical thinking, and personal artistic expression. The study further discusses the changing role of educators in balancing technological integration with the preservation of human imagination and design individuality. By adopting a human-centered perspective, the paper highlights the importance of maintaining emotional sensitivity, cultural understanding, and authentic creative processes in an increasingly AI-mediated educational environment. The research aims to contribute to the ongoing discourse on ethical and responsible AI integration in creative disciplines and proposes the need for pedagogical approaches that encourage collaboration between human intelligence and artificial intelligence rather than replacement. Ultimately, the study argues that technology should support, not overshadow, the human essence of creativity in fashion design education.
Maja Nikolova1; Djuradj Caranovic1
The development of modern technology and its application in everyday life point to the need to prepare young generations for a new era in which robotics will play an important role. Therefore, as part of the teaching process, robotics is increasingly present in our primary and secondary schools. Namely, the use of robot in teaching allows us to acquire new knowledge through the game, breaks the fear of the unknown in new technologies, develops imagination and creativity and to acquire new knowledge and skills in programming, mechanics and electronics. In addition, the construction and programming of robots represent a motive for further advancement in work as well as in school and out-of-school activities. The integration of artificial intelligence (AI) into education is transforming not only the physical presence of robots in the classroom but also the less visible processes through which AI determines what educational robots perceive, how they respond, and what they teach next. Over the past decade, AI in education has expanded from experimental applications to a wide range of technologies, including adaptive learning platforms, intelligent tutoring systems, learning analytics, and generative language models capable of interacting with students. This paper examines the growing role of AI-powered educational robots and considers both the empirical evidence of their impact on learning and the normative frameworks developed by international organizations. Particular attention is given to the distinction between evidence derived from empirical studies and recommendations concerning the responsible and ethical use of AI in education. By bringing these perspectives together, the paper highlights the opportunities offered by AI for personalized and interactive learning while emphasizing the importance of maintaining a clear distinction between demonstrated educational outcomes and broader institutional recommendations. The aim of this paper is to point out the need for teacher education to use educational robots in their work and AI for personalized and interactive learning. The methodology of work is based on the previusly published experience in this field as well as on the analysis of the offered facts. The results obtained by the qualitative analysis will point out the important place for robotic education in the new system of upbringing and education.
A conceptual framework for higher education
Rajni Bala1; Parul Kulshrestha1
Artificial intelligence (AI) has rapidly transformed higher education, creating unprecedented opportunities for teaching, learning, and research while simultaneously raising complex ethical, social, and governance challenges. Although AI literacy has emerged as a critical competency for university students, existing studies predominantly emphasize technical knowledge and AI adoption, with limited attention to integrating ethical reasoning, responsible behaviour, social responsibility, and institutional support into a unified framework. This conceptual paper addresses this gap by proposing the Responsible AI Literacy Development Framework (RAIL Framework) for higher education. Adopting an integrative literature review approach, the study synthesizes contemporary scholarly literature and international policy frameworks, including UNESCO, the European Commission, and the OECD, to identify the core dimensions of responsible AI literacy. The proposed framework conceptualizes responsible AI literacy as a developmental process beginning with institutional enablers including governance, policies, curriculum integration, and faculty capacity which foster students' AI foundations, cognitive competencies, ethical competencies, responsible AI behaviour, and social responsibility, ultimately leading to responsible AI literacy and the development of future-ready graduates. By integrating educational, ethical, and institutional perspectives into a single developmental model, the framework extends existing AI literacy research beyond technical competence and positions responsible AI literacy as a multidimensional capability essential for higher education. The paper contributes to theory by offering a comprehensive conceptual framework and provides practical implications for curriculum design, institutional governance, faculty development, and policy formulation aimed at preparing ethically informed, socially responsible, and AI-ready graduates in an increasingly AI-driven society.
Interdisciplinary perspectives, ethical challenges, and social responsibility in the digital learning ecosystem – a conceptual study
Anitha K1
The rapid integration of Artificial Intelligence (AI) into online education has significantly renovated the pedagogical landscape, reshaping teaching methodologies, learner engagement, assessment systems, and institutional decision-making processes. AI-driven educational technologies such as adaptive learning platforms, intelligent tutoring systems, automated evaluation tools, predictive analytics, and generative AI applications are progressively swaying the delivery and personalisation of digital learning experiences. This conceptual study discovers the developing role of AI in online education through the lens of educational innovation, socio-cultural transformation, and responsible technological adoption. The paper highlights key opportunities offered by AI in online education, including personalised instruction, learner analytics, scalability of education, and support for diverse learning needs. Simultaneously, it critically analyses emerging ethical challenges such as surveillance concerns, unequal access to AI technologies, transparency of algorithmic decisions, academic integrity issues, and the potential erosion of human-centred pedagogy. Additionally, the study proposes a conceptual framework for responsible AI adoption in online education based on principles of ethical governance, inclusivity, transparency, digital equity, and pedagogical accountability. The study contributes to the rising discourse on AI in education by offering a theoretically grounded understanding of how online education can balance technological advancement with ethical responsibility and social sustainability. The findings are expected to provide valuable insights for researchers, educational institutions, policymakers, and practitioners seeking to design equitable, human-centric, and socially responsible AI-enabled learning environments.
Layla Ajrouh1
Currently there is a big transformation in the educational environment due to the fast - evolving artificial intelligence (AI) landscape. How students are learning, interacting, accessing and using knowledge is different. With this new context, the mission of educators is becoming more than information transmission, instead, it is guiding, mentoring, fostering critical thinking and human-centered competencies to build global students able to operate in the new era. Guided by evolving research insights on contemporary educational practice, this chapter explores how educators can rethink their knowledge transferring roles to become key supporters of transformative learning. To benefit from the opportunities and cope effectively with the challenges brought by AI, the study highlights the importance of promoting lifelong learning skills. To achieve this, there is a need for a structured professional development for educators to help them ensure human interaction and healthy learning environments while integrating AI in their classes. Therefore, redefining educators' roles is essential for more engaged, personalized and transformative learning experiences and for preparing new generation of learners ready for the demands of the new digital future.
The CRAFT methodological framework for responsible design and reporting
Walid Allioua1; Ahmed Chaouki Hoadjli2
Recent developments in generative artificial intelligence (GenAI) have made it increasingly possible for educators and researchers to build customized AI tools without advanced programming knowledge. Through natural-language configuration, users can now define the role of an AI assistant, set interaction rules, restrict inappropriate outputs, provide knowledge sources, and shape task-specific learning support. This growing accessibility creates important opportunities for education, but it also raises methodological and ethical questions. If custom AI tools are used in teaching, assessment, or educational research, their design choices must be made visible, tested systematically, and reported transparently. This paper addresses the following question: how can custom AI tools in education be designed and documented in a responsible and methodologically rigorous way? To answer this question, the study proposes the CRAFT methodological framework, which organizes AI tool development into five connected stages: Conceptual Design, Realization, Assessment, Field Testing, and Transparent Evaluation and Reporting. The framework is developed through a conceptual synthesis of literature on GenAI in education, design-based research, educational methodology, and AI-mediated learning environments. The expected contribution is a practical and platform-independent structure that helps researchers and educators move beyond ad hoc AI use toward more transparent, comparable, and ethically responsible practice. By framing custom AI tools as educational and research instruments rather than neutral technologies, CRAFT supports responsible integration of GenAI into contemporary educational contexts.
Artificial intelligence in higher education between ethical governance, cognitive transformation, and social responsibility
Zbigniew Grzymała1; Dominik Baldowski2; Agnieszka Wójcik-Czerniawska1
The widespread adoption of generative artificial intelligence (AI) in higher education confronts universities with a tension between three orders: the cognitive (how AI transforms the processes of thinking and learning), the normative (how to regulate its use in an ethical and equitable manner), and the institutional (how the university, as an institution of public trust, should respond to these changes). This paper argues that the category binding these orders together is algorithmic literacy, understood not as a technical competence but as a condition for preserving human agency in a learning environment increasingly mediated by algorithmic systems. The paper is conceptual and review-based in character and draws exclusively on scholarly sources and documents issued by international organizations. The conclusions of this study are that the effects of artificial intelligence in higher education are determined not by the technology itself, but by the way it is embedded in teaching practice, in normative frameworks, and in institutional culture. Algorithmic literacy is therefore not merely another technical competence, but a condition for preserving human agency in an algorithmically mediated environment.
A sequential mediation model of attitude and perceived behavioural control
Oluwanife Segun Falebita1
The increasing integration of artificial intelligence (AI)-powered chatbots in higher education necessitates a deeper understanding of the factors influencing users' preferences for these technologies. Grounded in planned behaviour theory, this study examines the direct and indirect effects of technological readiness, attitude, and perceived behavioural control on chatbot preference. A sequential mediation model was developed in which attitude and perceived behavioural control explain how technological readiness (TR) translates into chatbot preference (CPR). Data were collected from 317 students in Nigerian higher education institutions and analysed using Partial Least Squares Structural Equation Modelling (PLS-SEM). Results show that TR significantly predicts attitude (β = 0.607, t = 17.242, p < 0.05) and perceived behavioural control (PBC) (β = 0.260, t = 4.405, p < 0.05), but not CPR directly (β = 0.053, t = 0.851, p = 0.395). Attitude significantly influences PBC (β = 0.377, t = 5.981, p < 0.05) and CPR (β = 0.483, t = 6.588, p < 0.05), while PBC also predicts CPR (β = 0.137, t = 2.395, p < 0.05). Indirect effects confirm significant mediation through attitude (β = 0.293, t = 5.878, p < 0.05), PBC (β = 0.036, t = 2.053, p < 0.05), and their sequential pathway (β = 0.031, t = 2.014, p < 0.05). Findings indicate that TR influences CPR primarily through attitudinal and control-based mechanisms rather than direct effects. Institutions should design AI-enhanced learning environments that strengthen users' technological readiness, positive attitudes, and perceived behavioural control to foster chatbot preference.
Institutional Support, Perceived Learning Outcomes, and the Perceived Social Value of AI
Ghofrane Barhoumi1; Jamel Eddine Gharbi2
The growing integration of artificial intelligence (AI) into higher education raises important questions about its implications beyond academic performance, particularly its potential to foster socially responsible attitudes among students. Drawing on Bandura’s social cognitive theory, this study examines how perceived institutional support, perceived learning outcomes, and students’ perceptions of AI-assisted learning relate to the perceived social value of AI in Tunisian higher education. Data were collected from 380 university students in Tunisia through a self-administered questionnaire and analyzed using partial least squares structural equation modeling (PLS-SEM) with SmartPLS 4. The measurement model demonstrated satisfactory internal consistency, convergent validity, and discriminant validity. The findings reveal that perceived institutional support positively influences students’ perceptions of AI-assisted learning (β = 0.075, p = 0.047), while perceived learning outcomes exhibit a substantial positive relationship with the social outcome construct (β = 0.477, p < 0.001). Mediation analysis further identifies significant indirect effects of perceived institutional support on students’ perceptions through perceived learning outcomes (β = 0.114, p < 0.001), and of perceived learning outcomes on the social outcome construct through students’ perceptions of AI-assisted learning (β = 0.170, p < 0.001). These findings highlight the importance of perceived educational benefits in connecting institutional support, AI-assisted learning experiences, and students’ socially oriented perceptions of AI. The study contributes to discussions on responsible AI integration and higher education marketing by emphasizing the role of institutional environments and perceived learning gains in shaping students’ perceptions of AI’s potential contribution to society.
Kunjumol P X1; Pratheesh Abraham2
Mathematical modelling is a cognitively demanding and educationally vital competency in secondary mathematics, requiring learners to translate complex real-world situations into structured mathematical forms and critically interpret their solutions. Conventional classrooms, however, rarely provide the reflective, disorienting, and meaning-making conditions that genuine modelling development requires. This paper introduces the AI-Supported Transformative Lesson Model (ATLM), a novel pedagogic framework that integrates Artificial Intelligence tools within teaching design, lesson planning, and feedback practices, grounded explicitly in Jack Mezirow's Transformative Learning Theory (TLT). The ATLM involves five pedagogic phases: Disorientation, Critical Reflection, Discourse, Reframing, and Integration, all supported through AI tools that scaffold, challenge, and respond to student thinking in real-time. AI-powered lesson planners design tasks to deliberately surface cognitive conflict, while intelligent scaffolding agents help students to critically review their assumptions. AI-driven formative feedback systems also provide tailored, process-level feedback to support reflective engagement with each stage of the mathematical modelling cycle (problem formulation, abstraction, representation, validation, and interpretation). The paper ends with the proposal of the AI-Supported Transformative Lesson Model, a conceptual framework that maps the intersection between AI-supported pedagogy and Transformative Learning Theory phases for building mathematical modelling skills. This model contributes to the broader conversation on transformative, subject-specific AI integration in secondary education, and offers a potential basis for future empirical validation and classroom application.
A comparative study of higher education institutions in India and Serbia
Rashid Hashmi1
The growing integration of AI-assisted learning platforms in higher education has transformed teaching, learning, and academic support systems across the world. While these technologies offer personalized learning experiences, real-time feedback, and improved access to educational resources, their effectiveness largely depends on the level of trust students place in such systems. Understanding the factors that shape student trust has therefore become essential for the successful adoption of AI-enabled educational technologies. This study examines student trust in AI-assisted learning platforms through a comparative analysis of higher education institutions in India and Serbia. Drawing upon technology acceptance, educational trust, and digital learning frameworks, the research investigates how perceived usefulness, transparency, data privacy concerns, learning effectiveness, and institutional support influence students' willingness to engage with AI-based educational tools. The study adopts an empirical approach using survey data collected from university students in both countries. Statistical analysis will be employed to identify similarities and differences in student perceptions across distinct educational and cultural contexts. The findings are expected to contribute to the growing discourse on artificial intelligence in education by providing evidence-based insights into the social, ethical, and pedagogical dimensions of technology adoption. The study also offers practical recommendations for universities, policymakers, and educational technology developers seeking to foster trust, enhance learner engagement, and promote responsible implementation of AI-assisted learning environments in higher education.
Personalized learning, adaptive assessment, and future pedagogical strategies
Saka Abdulganiyu1
This paper explores the transformative impact of artificial intelligence (AI) on contemporary educational paradigms. By integrating AI-driven technologies such as personalized learning systems and adaptive assessment tools, educators can tailor instructional content to meet individual student needs, thereby enhancing engagement and learning outcomes. The study examines emergent pedagogical strategies enabled by AI, addressing both opportunities and challenges associated with its implementation in diverse educational settings. Furthermore, the paper discusses future directions for AI integration, emphasizing its potential to revolutionize teaching and learning processes in the digital age. Through a comprehensive analysis, this research aims to inform educators, policymakers, and researchers about the transformative potential of AI in shaping the future of education.
Opportunities, challenges, and future directions
Gurinder Singh1; Anupma Rajesh1; Shipra Saraswat2; Vedant Patil1; Prakhar Chandra2
The education industry is rapidly evolving in the wake of Artificial Intelligence (AI), revolutionizing its method of instruction, learning, and how institutions function. As AI technologies are integrated into education, it has become possible to learn at our own pace in our own learning space. IT systems, automated assessment systems, virtual assistants and learning analytics are increasingly being used to boost student engagement, enhance learning outcomes and ease the burden on teachers. AI-powered learning platforms can adapt to the individual needs of each student, deliver targeted instruction, give immediate feedback and anticipate needs, all of which can promote student success. Further, AI can provide insights into student performance data to facilitate data-driven decisions in educational settings, thereby helping decision-makers to allocate resources effectively and identify students at risk. The advent of generative AI resources presents even more possibilities for creating content, learning together and gaining new skills related to the digital economy. While AI offers great promise in education, there are several challenges, such as concerns about data privacy, algorithmic bias, ethics, digital inequality, and role transformation for educators. To address these challenges, it is essential to have robust governance frameworks, ethical guidelines and capacity-building initiatives to ensure the equitable and responsible use of AI. This research paper will be discussing all the aspects, uses, advantages and disadvantages of the transformation of AI in the fields of education. It also elaborates on possible directions for future development of AI-based learning ecosystems that promote inclusivity, innovation and lifelong learning. The results underscore the promise of AI in designing more inclusive, effective, and student-centric learning spaces and the need to consider the ethical implications of technology.
Generative Artificial Intelligence in Education from the Perspective of Students in Serbia
Dobrinka Kuzmanović1; Tijana Milošević1
In recent years, tools based on generative artificial intelligence (GenAI) have increasingly shaped the digital environment of children and young people. Accordingly, this paper aims to explore the experiences of students in Serbia in using GenAI tools in educational contexts. Particular attention is paid to the types of tools young people use, the frequency and purposes of their use, as well as the existence of school rules, potential misuse, and its consequences. The paper also examines how students assess their own GenAI literacy and what opportunities they identify for developing it within formal education. The paper integrates findings from quantitative and qualitative research conducted in 2025 as part of the international EU Kids Online study. The online survey was conducted on a representative sample of 1,675 students aged 9–16 from 64 schools, while individual interviews were conducted with 16 students aged 13–17. The results show that GenAI tools are most commonly used for educational purposes, primarily for writing essays (43% of surveyed students) and summarising or explaining longer texts (30%). Students occasionally misuse these tools, particularly when completing homework assignments and tests. Teachers’ attitudes range from prohibiting to encouraging their use, while, according to students, misuse often goes without a response from teachers. Most respondents believe that using GenAI does not require any specific skills, suggesting a limited understanding of these technologies. At the same time, most students have not had an opportunity to learn about GenAI within formal education. The findings point to the need to develop educational policies and practices that support the safe, responsible, and pedagogically meaningful use of GenAI.
Oluwatomisin Marvellous Ogundeji1
This study modeled metacognitive and motivational characteristics on secondary school students' conceptual change in light. This study employed a model-based correlational survey research design. The sample size was 384 SS3 Students taking Physics, which was determined using the Taro Yamane formula, but selected using a multi-stage sampling procedure. The students' Metacognitive Characteristics Inventory (SMCCI), the Students' Motivational Characteristics Inventory (SMCI), and the Light Phenomena Concept Test (LPCT), were used for data collection. The estimate of internal consistency for the SMCCI, SMCI, and LPCT was determined using Cronbach's alpha. The internal consistency indices obtained were 0.91, 0.84, and 0.86, respectively. The collected data were analyzed using RStudio version 4.4.3. Specifically, the SEMinR package and plspm. R is a statistical foundation of AI, suited for developing and validating AI models. R contains packages that implement machine learning algorithms that AI often depends on. The research questions were answered using path coefficients (β), effect size (f²), and coefficient of determination (R²). The hypothesis was tested using a bootstrapping procedure. The findings of this study revealed the most parsimonious path model that explains the effects of metacognitive and motivational characteristics on students' conceptual change in light. The metacognitive and motivational characteristics also had significant predictive power for students' conceptual change in light. It implies that, if the students' metacognitive and motivational characteristics are not considered in designing conceptual change instruction for students taking Physics in senior secondary school, students' conceptual change may be hampered, thus promoting poor performance in the subject.
A systematic review
Malathi Letchumanan1
Artificial intelligence (AI) chatbots are increasingly being integrated into mathematics learning. While previous reviews have explored the impact of chatbots on education, limited attention has been given to how mathematics students specifically utilise these tools for learning. To address this gap, this study systematically reviewed 26 articles published between 2020 and 2026, following the PRISMA guidelines, to examine the benefits, and challenges associated with the use of chatbots in mathematics learning. The findings suggest that AI chatbots have considerable potential to enhance both cognitive and affective learning outcomes. Nevertheless, concerns remain regarding ethical considerations and the accuracy, reliability, and suitability of chatbot-generated solutions to mathematical problems. The insights derived from this review provide valuable recommendations for practitioners, policymakers, and chatbot developers.
Arshad Bhat1
Incorporating AI into educational systems has drastically altered the landscape of teaching design, lesson planning and the feedback process in today's learning environment. Teachers are increasingly pursuing and utilizing AI tools for crafting individual lesson plans, creating personalized lesson content, checking student progress, and giving immediate feedback. Such technological developments have improved efficiency, accessibility, engagement of students and provided opportunities for data driven and student centred education. The paper highlights the potential of AI-powered tools to enhance pedagogical practices and enhance instructional management. It will explore the application of intelligent systems to the support of the teacher in curriculum planning, assessment of the classroom, differentiated instruction and automated generation of feedback. The study also explores the possible uses of AI for inclusive education, encompassing different needs of learning, and the provision of personalized learning experiences. The paper also explores current themes of overuse of technology, algorithm accuracy issues, less human interaction, issues associated with privacy, and accountability challenges in educational decisions. The paper contends that Artificial Intelligence (AI) can serve as a facilitative teaching aid, not a teacher. Emotional understanding, creativity, ethics and social interaction are essential to effective teaching and are key human capacities. For educational systems to successfully incorporate AI, it's equally important for technological advancements to work closely with educational methods, or pedagogy. The study demonstrates that the proper use of AI-enhanced teaching methods has the potential to enhance teaching, learning, and education in a flexible, efficient, and inclusive way without losing its human essence.
Offia Tugwell Owo1; Sylvester Akujuru1
This study investigated the effects of artificial intelligence (AI) integration in pedagogical instruction on the academic achievement of students in Power Electronics at Ignatius Ajuru University of Education, Port Harcourt, Rivers State, Nigeria. A quasi-experimental design with a pre-test and post-test control group arrangement was adopted. The population of the study comprised 124 students of Electrical/Electronic Technology Education option in the Department of Technical Education. A purposive sampling technique was used to select 62 students as the sample of the study, who were assigned equally to an experimental group (n = 31) and a control group (n = 31). The experimental group received instruction through AI-integrated pedagogical methods, while the control group was taught using conventional instructional approaches. A researcher-developed Power Electronics Achievement Test (PEAT) was used as the instrument for data collection. The instrument was validated by three experts and had a reliability coefficient of 0.82 using Kuder-Richardson Formula 20 (KR-20). Data collected were analysed using Analysis of Covariance (ANCOVA). The findings of the study revealed a statistically significant difference in the academic achievement of students taught with AI-integrated pedagogical instruction compared to those taught through conventional methods (F(1, 59) = 47.31, p < 0.05). Consequently, the study concluded that AI integration in pedagogical instruction substantially improves students' academic achievement in Power Electronics. It was therefore recommended among others that lecturers in Electrical/Electronic Technology Education programmes should be trained and supported to embed AI-driven instructional tools into their teaching practices.
BloomBot, VelikovskyBot, evolutionary culturology GPTs and EthiSizer GPTs as governed epistemic interfaces
Joseph Tesla Velikovsky1; Ryan T Dean2
Archive-grounded generative AI systems are increasingly important for science, education, culture, and public-interest knowledge work, yet their reliability depends on more than model performance alone. This paper examines three related systems - BloomBot, VelikovskyBot, and a family of Evolutionary Culturology GPTs and EthiSizer AI GPTs - as case studies in transparent, socially responsible human-AI-archive interaction. We argue that such systems should be understood not merely as chatbots, but as governed epistemic interfaces: systems that mediate between users and curated bodies of expert knowledge, while making sources, limits, and interpretive steps visible. BloomBot, developed by Ryan T. Dean, supports engagement with Howard Bloom's interdisciplinary archive. VelikovskyBot extends this approach to J. T. Velikovsky's archive and the meta-meta-science of Evolutionary Culturology. The Ev Cult GPTs further explore how archive-grounded AI can assist interdisciplinary education and public reasoning by mapping concepts across domains through the universal unit of the fractal HOLON/parton. The paper also considers the international portability of such systems, comparing their potential relevance for Australian, North American/Canadian, Balkan/EU, and MENA/Global South educational and cultural contexts. We propose a practical design framework for public-interest AI: source grounding, inference labelling, persona restraint, epistemic humility, and user-facing explainability. These principles help reduce hallucination, preserve intellectual context, and build public trust in AI-assisted inquiry.
R. Vijayakumar1; S. Mohamed Moosa2
Artificial Intelligence (AI) has become one of the most transformative forces of the 21st century, reshaping industries, services, and everyday life. In education, AI represents a paradigm shift rather than a mere technological add-on, redefining how knowledge is delivered, assessed, and experienced. By leveraging tools such as machine learning and natural language processing, AI enhances teaching and learning processes globally. The advantages of AI integration are substantial. Automated grading and administrative tasks allow teachers to focus on mentoring and creativity. Assistive technologies such as voice-to-text, screen readers, and translation tools break language barriers and empower students with disabilities. AI-powered remote learning platforms extend education to rural and underprivileged areas, while analytics help institutions refine curricula and student support systems. Virtual tutoring, interactive lessons, and instant feedback further personalize learning experiences (UNESCO, 2023). Nevertheless, challenges persist. Data privacy must be safeguarded to protect student information. Algorithmic bias risks perpetuating inequality if systems are trained on flawed datasets. Over-reliance on technology may reduce human interaction, empathy, and creativity. The digital divide continues to hinder underfunded schools, and many educators require training to effectively integrate AI. Policymakers must establish clear guidelines to ensure equity, accountability, and transparency (OECD, 2022). India's National Education Policy 2020 emphasizes introducing AI at the school level, highlighting the need for reliable internet, well-equipped labs, teacher training, and support for regional languages. Ultimately, AI should be viewed as a supplement rather than a substitute for human interaction. When adopted responsibly, it can revolutionize education by making learning more accessible, engaging, and cost-effective, while empowering educators to focus on creativity and mentorship
Hana Karaahmetović1
Contemporary education increasingly aspires to a learner-centred model, which requires a high degree of personalised learning, particularly when working with children who follow individual education plans (IEPs). Creating materials suited to students with different levels of ability is a demanding and time-consuming task for primary school teachers. This paper examines the pedagogical potential of artificial intelligence tools as digital assistants that can support teachers in adapting instructional content to the specific needs of individual students. The focus is on exploring how AI tools can assist teachers in transforming standard lessons and differentiating instructional content. The paper examines their potential for creating personalised materials, from simplifying texts and identifying key information for students at IEP1 and IEP2 levels to designing additional tasks for gifted students at IEP3 level. It also considers the advantages of using AI tools for the technical preparation and design of teaching materials, as well as their potential to reduce teachers’ workload and allow them to devote more time to direct pedagogical work with students. This area represents an emerging development in the continuing transformation towards more inclusive education. Further research is needed to gain a fuller understanding of the pedagogical benefits and long-term effects of using artificial intelligence in teaching.
A pedagogical approach
Nemanja Spasenovski1; Karlo Bojičić2
A growing number of research on artificial intelligence(AI) in education focuses predominantly on questions of how AI tools can be effectively integrated into curricula, how they personalize learning, and how they improve measurable outcomes od educational process. Research on implementation and personalization addresses the question of how AI can be used in education, implicitly assuming that its mere presence in educational contexts confirms its pedagogical purpose. This paper argues that this assumption requires critically pedagogical examination because the mere tehnical benefits of AI do not constitute grounds for a pedagogical relationship, nor do they guarantee orientation toward the child's development. Drawing on Erich Fromm's critical distinction between having and being, this paper frames the encounter between the child and algorithmic AI environments as one in which the child is systematically reduced from a subject in development to a set of data optimized for engagement and consumption. This reification, rooted in the structural logic of market rationality, creates conditions Fromm identifies as generative of malignant aggression, or destructiveness that emerges not from biology but from alienation from one's human essence. In that sense purposelessness of AI is fundamental question, as it enables environment devoid of human meaning and pedagogical purpose. Drawing on cognitive script theory (Huesmann) and social learning theory (Bandura), we argue that children encode behavioral patterns from any consistent model regardless of intent and purpose. The absence of intentionality therefore does not diminish the pedagogical risk but it is the pedagogical risk itself. This paper argues that the appropriate response is a pedagogical one, not in the sense of specific methods or tools, but in the foundational sense. The pedagogical approach to AI establishes the formative conditions under which the child becomes the subject who gives purpose to AI, rather than the object shaped by it. It is precisely this foundational reconsideration that existing research on AI in education has yet to undertake, and that this paper seeks to initiate.
Omar Abdulrhman Sheikh Najeeb1; Ahmed Abu Hajar2
This research aims to examine how effective modern teaching methods in teaching the field of Applied Chemistry. To achieve this, data and students' opinions were collected from first- and second-year male and female students at Maaref University of Applied Sciences in Syria. The study also includes previously published research results from the university journal, covering the period from the first quarter of 2024 to the second quarter of 2026 in the field of Applied Chemistry. In addition, the research discusses students' acceptance of studying the specialization in English, while taking into account suitable class sizes in classrooms and laboratories. It also highlights the importance of field training in several factories, such as (Riva Pharma for pharmaceutical production and Ryan Food for feeding products…) This teaching approach is considered the first of its kind in the Syrian Arab Republic.
ThinkAI is artificial intelligence use thinking partner approaches
Raihand Ramadhani Abdul Sayeed1; Fadly Mutadhid Ahzam2
Nowadays, the Artificial Intelligence (AI) industry has been implemented in every sector such as education, business, economy, etc. According to McKinsey and Company research, 88% of industries have implemented AI [1]. Moreover, the leverage of Artificial Intelligence in education, as highlighted by Prof. Janice Kay, CBE, shows that 85% of students access AI worth 88% [2]. The challenges that emerge in the education sector include students having lower confidence, lower ability to make decisions, and cognitive loading. Research by the University of Texas indicates that AI dependency in education can activate a passive mindset, where the brain only accepts results without carrying out an in-depth analysis process [3]. ThinkAI emerges as a partner for students in higher education, it can serve as a companion to finding answers based on multiple points of view. For example, when students need to answer an essay, ThinkAI helps by providing answers from different points of view. ThinkAI uses a Thinking Partner approach, which focuses on human-AI interaction where AI provides various perspectives in answering questions. This research uses a Systematic Literature Review to gather references from journals, books, and other sources to identify, evaluate, and synthesize.
Jorge Isaac Torres Manrique1; Shailja Vasudeva1
The application of information and communication technologies and artificial intelligence in the education system presents various advantages, disadvantages, and challenges. The current reality is unfolding against the backdrop of new scenarios and perspectives. In this regard, in this article, the authors explore and develop this crucial and far-reaching topic, concluding that: i) The application of new technologies and artificial intelligence represents a significant opportunity to strengthen the education system. It is important to recognize that we are facing a new reality; ii) The demonization and rejection of the application of new technologies and artificial intelligence are not wise approache, iii) Education is being transformed by the advent of new technologies and artificial intelligence. Specifically, with regard to digital immigrants and digital natives, iv) the fundamental rights of digital natives cannot be ignored or postponed. And thus, they conclude by proposing: i) The adoption of public policies to raise awareness of the importance of applying new technologies and artificial intelligence to the education system, ii) It is essential to implement public policies for training teachers and students on the nature and scope of new technologies and artificial intelligence within the education system, iii) Properly adapt the application of new technologies and artificial intelligence within the education system in accordance with the nature and requirements of digital natives; iv) Establish legal limits to safeguard and ensure the effective recognition of the fundamental rights of digital natives.
Transforming information services, user experience, and digital learning
Sudeep D K1; Madan D D2; Prakash3; Ashok Kumar B S1; Disha N S1
The integration of artificial intelligence (AI) technologies is revolutionizing academic libraries through innovative services that transform user experience and digital learning environments. Technologies such as generative AI, intelligent discovery tools, chatbots, recommendation engines, and metadata generation using natural language processing are transforming academic libraries by improving information services and digital scholarship. This research paper aims to explore the role of AI in shaping academic library services and the impact it creates. It adopts a descriptive and analytical research approach based on contemporary academic literature related to AI in libraries and information sciences. The analysis shows that academic libraries have embraced AI technologies in areas such as personalized information seeking, virtual reference services, learning analytics, and research assistance. The use of AI technologies in academic libraries enhances information access and efficiency, improves information literacy, and transforms learning environments. In addition, academic libraries face some ethical and professional challenges when integrating AI technologies because of the potential risks involved in data privacy, bias in algorithms, intellectual freedom, and digital inequality. Also, there is need for professional development among library personnel to ensure responsible adoption and implementation of AI technologies. While the adoption of AI technologies presents great opportunities, it should involve consideration of professional ethics and responsibilities. The results demonstrate that academic libraries are transforming into intelligent and knowledge-oriented ecosystems. Academic libraries are increasingly embracing the use of AI technologies to enhance information access and digital learning practices.
A bloom-to-fink pedagogical transition
Archana Jog1
The development and growth of online learning in higher education has increased the need for pedagogical frameworks to support deeper student engagement and meaningful learning experiences. Bloom's taxonomy has been a guide for a long time for shaping of learning objectives in management education; however, multidimensional nature of learning in digital classrooms is not fitting completely into the hierarchical cognitive processes of the taxonomy. This study explores how integrating Fink's Taxonomy of significant learning can shape the design of an online post graduate management course. The research is based on survey responses from 186 postgraduate students and 24 faculty members involved in an online management course delivered through a webinar-based teaching environment. The study explores how course activities aligned with Fink's dimensions of learning—including application, integration, human dimension, caring, and learning how to learn—shape students' engagement and meaningful learning outcomes. The results show that the activities built around reflection, real-world application, and personal relevance help students engage more actively and understand course concepts better in large online classrooms. The study contributes to the scholarship on active learning in higher education by demonstrating Fink's taxonomy applied as a practical framework for designing meaningful learning experiences in online management education.
Rethinking student agency and ethical learning practices
Mohand Ouamar Akab1; Abdelkader Kermezli2
This study explores the impact of Generative Artificial Intelligence on student agency and language learning practices in the Algerian higher education context. It focuses on Algerian EFL learners' attitudes toward the use of generative AI and examines how human–AI collaboration is reshaping educational experiences, learner autonomy, and ethical engagement with digital technologies. The study involves EFL students from different Algerian universities. In Algerian EFL classrooms, language learning is still largely based on traditional teacher-centered approaches, which often limit opportunities for interaction and autonomous learning. Generative AI technologies, however, provide learners with personalized feedback, adaptive learning experiences, and interactive language practice that may enhance engagement and independent learning. At the same time, their growing presence in education raises ethical concerns related to academic integrity, overreliance on AI tools, digital inequality, and responsible use of technology. Using a mixed-methods approach based on semi-structured surveys, this study investigates the extent to which Algerian EFL learners use generative AI for language learning, their perceptions of ethical human–AI collaboration, and the challenges influencing on effective implementation. The findings reveal that although the use of generative AI remains limited due to infrastructural constraints such as weak internet connectivity and insufficient technological resources, EFL students still hold positive perspectives toward its educational potential. The study concludes that generative AI can foster educational transformation in Algeria by promoting learner autonomy, digital literacy, and ethically informed human–AI collaboration in language education.
Ethical considerations and competency development
Elomari Driss1
This paper explores the transformative potential of Artificial Intelligence (AI) within pedagogical engineering, focusing on its application in designing and optimizing digital learning environments. Drawing upon extensive research in e-learning, MOOCs, and pedagogical innovation, this contribution will delve into how AI-driven tools and methodologies can enhance learning experiences, foster competency development, and facilitate effective formative evaluation through e-portfolios. We will discuss practical implementations and theoretical frameworks for integrating AI into educational design, while critically examining the ethical implications and social responsibilities associated with these advancements. The presentation will highlight strategies for developing essential soft skills in learners within AI-augmented educational contexts, addressing the challenges and opportunities for institutional change. This interdisciplinary perspective aims to provide insights for educators, researchers, and policymakers on harnessing AI to create more inclusive, adaptive, and impactful digital learning ecosystems.
Reimagining pedagogical practice in the age of artificial intelligence a theoretical framework for technology-enhanced learning in tertiary education
Boutkhil Guemide1
The rapid integration of artificial intelligence (AI) into higher education has necessitated a fundamental rethinking of pedagogical frameworks. Traditional instructional models—largely conceived before the proliferation of digital and intelligent technologies—are increasingly inadequate for addressing the complexities of contemporary learning environments. This paper introduces the Pentagonal Integration Model (PIM), a novel theoretical framework designed to guide pedagogical practice in technology-enhanced tertiary education settings. The model identifies five interdependent dimensions: the Teacher, the Learner, New Technologies and AI Applications, Content, and Teaching Methodologies. Drawing on constructivist and connectivist learning theories, as well as empirical research in educational technology and AI-assisted instruction, the PIM offers a holistic architecture for understanding how these five dimensions interact synergistically to produce enriched learning outcomes. The paper discusses the theoretical underpinnings of each dimension, provides a conceptual diagram of the model, and explores implications for curriculum design, faculty professional development, and institutional policy. The Pentagonal Integration Model represents a significant contribution to the literature on twenty-first century pedagogy, offering researchers and practitioners a robust lens through which to examine, evaluate, and redesign teaching processes in the age of artificial intelligence.
Student perspectives at a Global South university
Md Tariqul Islam1
The rapid proliferation of generative artificial intelligence (AI) tools has transformed the landscape of higher education, including general education (GenEd) courses. This study investigates how students at a Global South university engage with AI in GenEd courses, exploring the purposes, practices, challenges, and ethical considerations associated with its use. Utilizing a mixed-methods approach, data were collected through questionnaires incorporating both structured and open-ended questions, focusing on six key areas: awareness and familiarity with AI, the nature and extent of AI use in GenEd courses, motivations for using AI, challenges and risks encountered, perceived impact on learning, and perspectives on ethical and effective AI use. The study highlights a nuanced understanding of student motivations, ranging from efficiency and curiosity to academic performance enhancement, and underscores the potential of AI to both support and disrupt learning processes. These insights contribute to the growing body of research on AI integration in higher education, particularly in Global South contexts, where infrastructural and pedagogical factors may shape AI adoption differently from the Global North. The findings also carry implications for institutional policies, curriculum design, and teaching practices, advocating for clear guidelines, ethical frameworks, and faculty support to maximize the benefits of AI while mitigating its risks.
Hitesh Mohapatra1
This paper investigates the role of artificial intelligence (AI) in reshaping education, pedagogy, assessment, and institutional cultures, viewed through interdisciplinary lenses from education, sociology, and ethics. It analyzes how AI driven tools are reconfiguring teaching learning practices, automating and redefining assessment strategies, and driving broader socio institutional transformations. The discussion highlights both opportunities such as personalized learning pathways, adaptive assessment, and improved administrative efficiency and pressing ethical concerns, including algorithmic bias, data privacy, surveillance, and the erosion of human agency in educational decision making. Building on these insights, the paper proposes an ethical reflexive framework for socially responsible AI integration in education. It emphasizes the need to align technological change with pedagogical values, equity, and democratic participation at classroom, institutional, and policy levels. Concrete recommendations are offered for educators, academic leaders, and policymakers to embed transparency, accountability, and cultural sensitivity into AI supported pedagogy, assessment design, and institutional governance. The analysis underscores that AI driven transformations must be guided by social responsibility, not merely by efficiency or technological determinism.
Mode 4 knowledge production, innovationology, and the reconfiguration of socially responsible higher education in the Global South
Pitshou Moleka1
Contemporary debates on artificial intelligence are predominantly framed within technocratic, computational, and market-centered paradigms that frequently marginalize ethical plurality, community knowledge systems, spirituality, and culturally embedded understandings of human flourishing. This paper argues that the accelerating integration of artificial intelligence into higher education demands not merely technological adaptation, but a profound epistemological and civilizational reconfiguration of knowledge production itself. Drawing on the emerging paradigms of Mode 4 Knowledge Production and Innovationology, the paper develops a transdisciplinary framework for rethinking artificial intelligence within socially engaged and ethically grounded higher education systems, particularly in Global South contexts. Unlike earlier models of knowledge production centered primarily on disciplinary science, institutional research, or university–industry collaboration, Mode 4 emphasizes participatory, spiritually informed, culturally situated, and community-anchored co-creation of knowledge. Innovationology further expands this perspective by conceptualizing innovation as a complex, relational, and context-sensitive phenomenon shaped by interactions among institutions, technologies, ethical systems, local cultures, and marginalized communities. The paper contends that higher education institutions — especially faith-based and socially embedded institutions in Africa — possess unique potential to cultivate alternative AI futures grounded in social responsibility, cognitive justice, human dignity, ecological consciousness, and transformative community engagement. Rather than treating AI solely as a technological instrument, the study proposes its reconceptualization as part of a broader ethical and societal ecosystem requiring plural epistemologies, inclusive governance, and transdisciplinary collaboration. Ultimately, the paper advances a Global South-oriented framework for ethically grounded artificial intelligence capable of contributing to more humane, inclusive, and socially transformative futures.
Evidence from Sri Lanka's higher education context and implications for the Global South
Chammika Mallawaarachchi1
The use of artificial intelligence (AI) in higher education has changed the roles of academic libraries. Libraries now play a key part in encouraging AI literacy among students, staff, and stakeholders. Globally, 67 per cent of libraries are exploring or using AI tools, up from 63 per cent in 2024. Most libraries are still in early evaluation stages, and there are big differences between high-income and Global South contexts (Clarivate, Pulse of the Library, 2025). Sri Lanka is a good example. The World Bank classifies Sri Lanka as lower-middle-income (2025). Sri Lanka's AI Sri Lanka 2028 policy aims to build AI expertise among citizens, practitioners, decision-makers, and specialists (Ministry of Digital Economy, 2024). Academic libraries have an important but underexplored role in this scheme. This paper studies how Sri Lankan academic libraries can promote AI literacy within their institutions. A review of current literature, national policy, and institutional practices finds three main barriers: staff skills mismatches and reluctance to change; limited coverage of critical and ethical AI in Library and Information Science curricula; and organisational constraints typical of lower-middle-income education systems. These outcomes are set within Global South scholarship on fair AI adoption. This article highlights the need for context-sensitive, locally tailored strategies rather than direct models from high-income countries. It ends with evidence-based recommendations for curriculum reform, staff development, and policy changes in South and South-East Asian library systems.
Cultural transformation and ethical challenges
Mehak Jonjua1
Artificial Intelligence and metaverse technologies are quickly converging and changing the educational environment in many countries. South Asia is in a precarious position because it has over 600 million students who can take advantage of these digital technologies, but it also faces some major cultural issues. This research aims to explore the role of AI and the metaverse in transforming education in South Asia, with a particular focus on the countries of India, Pakistan, Bangladesh and Sri Lanka. This study has used a mixed-methods approach to understand how these technologies are changing educational practices in South Asia while taking into consideration the cultural and socio-economic context of each nation. The researchers will be using qualitative data obtained through interviews with educators, policymakers and students, and quantitative data obtained from technological adoption measures collected from urban and rural educational institutions. The researchers' analysis indicates that although AI personalization and metaverse platforms are positively affecting learner engagement and accessibility, they also have the potential to create or exacerbate existing digital divides, erode culturally relevant ways of learning and teaching, and create unresolved issues surrounding data privacy and algorithmic bias. The researchers also found that ethical ethical frameworks guiding the use of educational technology in South Asia are sorely lacking. These findings have great significance for policymakers, curriculum designers and technology providers, and call for collaboration among all stakeholders to create culturally responsive, ethical AI and metaverse-based educational environments that honor South Asia's rich cultural heritage and prepare the region's youth for an increasingly digital world.
Ethical innovation, academic integrity, and Human-Centered pedagogy in the digital age
Mohit Singh1; Balwant Singh2
In the realm of higher education, Artificial Intelligence (AI) is making significant strides in the fields of adaptive learning systems, automated evaluation tools, intelligent tutoring systems and data-driven educational management.AI is reshaping the landscape of higher education with the adaptation of AI-led systems for learner personalisation, automated assessment platforms, intelligent tutoring, and data-powered education management. AI has many promising potential benefits for personalised learning, accessibility, and institutional efficiency but also several key ethical and social issues arising: issues with academic integrity, algorithmic bias, issues with data privacy, issues with surveillance, issues with digital inequality. The paper critically evaluates the presence of AI in revising teaching, learning and assessing processes in an interdisciplinary and human-centred lens. The study adopts an international comparative approach using AI emerging practices in HED in Canada, EU, the Middle East and North Africa (MENA) and the Global South. It identifies and discusses the role of the differences in technological infrastructure, governance frameworks, digital literacy, and educational policy in shaping responsible embedding of AI in university environments. The role of teachers in the use of AI in learning classrooms is also evolving, and the paper highlights the need to maintain the critical thinking competency, creativity, and meaningful human interaction needed in classrooms where learning is digital. The study suggests an ethical and socially responsible framework for using AI in education, focusing on transparency, inclusiveness, pedagogical balance, and accountability.
Opportunities, risks, and future directions
Shivnath Ghosh1; Olivia Jana1
Generative Artificial Intelligence is advancing rapidly and revolutionising the STEM education practice from conventional to interactive, personalised and technology-driven practices. Generative AI-powered tools are likely to be used more and more in the development of automated content creation, intelligent agents that act as tutors, virtual laboratories, real-time problem solving through natural language interaction with computer-moderated learning environments. From enabling students to become more engaged, to simplifying difficult science and math concepts, these innovations have the potential to enhance learning by providing reliable and accessible education for students of all kinds. While generative AI offers many potential benefits to enhance learning in STEM education, its use also poses challenges concerning accuracy, ethics, privacy and security of student data, misuse or abuse of academic integrity policies in different disciplines and overreliance. This is the way we made accusations about improper use of AI tools in assessments, no transparency from AI on decision-making affects students critical thinking and poses challenges for the educational facilitators and institutions. As a result, the reconciliation of innovative approaches with responsible deployment is an important topic in contemporary education systems. This paper discusses the main opportunities, risks and future possibilities of generative AI in enhancing STEM education systems. This study synthesizes how AI can increase personalisation in learning and inter-disciplinary innovation while stressing the necessity for ethical guidelines, human oversight and acceptable practices of AI. The paper also describes a path forward to integrating generative AI into education while ensuring technological effectiveness and meaningful learning outcomes.
Oluseye Olugboyega1; Moses Adeyemi1; Abiola Odunayo Ajayi1; Opeyemi Adepeju Ojo1
This study investigated the extent to which an AI teaching assistant enhances planning, problem solving, decision making and logical reasoning among 72 construction management students using a pre-test and post-test quasi experimental design. Students completed eight cognitively demanding and discipline specific assessment tasks covering construction planning, managerial problem solving, decision analysis and critical reasoning. Parallel test forms were administered before and after structured exposure to an AI teaching assistant integrated into learning and practice activities. Descriptive analysis, paired samples t tests, Wilcoxon signed rank tests, repeated measures ANOVA and repeated measures MANOVA were used to evaluate baseline trends, statistical attribution of improvement and overall multivariate cognitive change. Practical educational impact was quantified using Cohen's d and partial eta squared. Gain score analysis and machine learning models were further applied to predict learning improvement, identify key cognitive predictors, detect learner profiles and classify students by improvement level. Results show statistically significant and practically large gains across all four competencies, with balanced improvement patterns and a very strong overall multivariate effect. Decision making and logical reasoning demonstrated the largest improvements and were the strongest predictors of total cognitive gain. Distinct learner profiles were identified, indicating heterogeneous learning responses. The findings provide robust empirical evidence that AI supported learning can strengthen integrated professional cognition in construction management education and inform adaptive instructional design.
Innovation, anxiety, and ethical responsibility
Chahrazad Laleg1; Kouider Makhlouf2
Generative artificial intelligence is revolutionising higher education by redefining learning, teaching, and academic integrity. AI-based tools such as ChatGPT can support students by explaining complex concepts, generating ideas, and providing rapid feedback. They can also assist educators in developing learning activities, creating instructional materials, and tailoring content to meet diverse student needs. Hence, generative AI presents important opportunities to advance pedagogical innovation. This paper studies generative AI in higher education through the interrelated dimensions of innovation, anxiety, and ethical responsibility. Using evidence from scholarly literature and institutional sources, it contends that universities should adopt a balanced approach rather than relying on simplistic responses such as total prohibition or unrestricted adoption. The central concern is not whether generative AI can be used in higher education, but rather how it can be integrated responsibly without compromising meaningful learning. The paper also highlights the importance of establishing clear institutional policies, promoting AI literacy for students and educators, strengthening transparent disclosure of AI practices, and designing assessment methods that prioritise reasoning, learning processes, and critical reflection. Lastly, the paper concludes that ethical and accountable integration of generative AI requires a balanced framework that recognizes its educational potential while addressing the ethical risks it introduces.
S.Janaki1; S.Sujatha2; R.Rohini3
Foundational literacy and Numeracy refers to the essential basic skills of reading, writing and mathematical reasoning that a child must acquire by the end of Grade 3.NIPUN Bharat and NEP 2020 were introduced to prioritize FLN because it is the baseline for lifelong learning. Without these basic skills, children struggle to catch up in later grades, which significantly increase the risk of drop outs. Even when they reach secondary classes and senior secondary including collegiate education this learning gap exists. Globally, 53 percent of 10 year olds in low and middle income countries cannot read and understand a simple story. According to NEP 2020, over five crore children in India's elementary schools lack foundational literacy and Numeracy (FLN) skills. This challenge gets widened when students reach upper primary classes and secondary classes. The present study will examine the existing gap in selected schools for upper primary classes i.e. 6, 7 and 8th grades and intervention strategy will be taken up including the components of oral language development, decoding, fluency in reading, reading comprehension, and writing pertaining to foundational literacy and key components of foundational numeracy will include pre-number notions, numbers and operations, shapes and geometrical understanding, measurement and data handling. Experimental study will be taken up including AI intervention pedagogy and assessment for enhanced institutional set up in the age of generative AI. The study will probe into existing challenges in teaching and learning of English language and mathematics in upper primary classes and suggest innovative value based pedagogy.
A human-centric framework for responsible agentic AI in digital TVET
Rama Yusvana1
Digital TVET is now expected to prepare learners for AI-mediated workplaces across rapidly changing professional and industrial sectors, yet many institutional responses still treat artificial intelligence as a tool for automation rather than a supervised support system for human development. This gap is visible in practice: students encounter AI tools without structured guidance, while educators manage implementation without clear ethical or pedagogical frameworks to draw on. This conceptual paper proposes a human-centric framework for responsible Agentive AI in Digital TVET, with specific attention to learning support, student readiness, teacher agency and ethical governance. The framework is informed by an ongoing research programme examining student readiness and AI integration challenges across Malaysian TVET settings. It positions Agentive AI as an LMS-integrated learning companion capable of providing guided hints, formative feedback, practice prompts, reflective questioning and learning-path recommendations within boundaries set by educators and institutions. Rather than replacing teaching judgement, the model defines AI as a pedagogical layer that strengthens self-directed learning while keeping teachers responsible for curriculum alignment, intervention decisions and assessment integrity. The framework integrates six interrelated dimensions: digital and AI literacy, equitable technology access, pedagogical alignment, student agency and well-being, explainable learning analytics and institutional safeguards covering privacy, bias, transparency and responsible use. The paper offers TVET institutions a practical pathway from ad hoc AI use toward structured, ethical and learning-centred implementation. Its broader contribution frames AI adoption in TVET as an institutional commitment to educational quality, inclusion and social responsibility, rather than a purely technical or operational upgrade.
Rethinking teaching, assessment, academic integrity and institutional transformation in the digital age
Obi Adaora Irene1
The rapid emergence of Artificial Intelligence (AI), particularly generative AI technologies, is transforming educational systems and prompting a fundamental reconsideration of pedagogical practices, assessment mechanisms, and institutional structures worldwide. While existing scholarship has extensively examined the technological and administrative dimensions of AI adoption in education, limited attention has been devoted to its implications for historical pedagogy within African higher education. This study critically examines how AI is reshaping the teaching and learning of history, transforming assessment practices, challenging conventional understandings of academic integrity, and influencing institutional transformation in African universities. The study adopts a critical integrative literature review methodology, synthesizing contemporary interdisciplinary scholarship from the fields of artificial intelligence, history education, digital humanities, educational technology, and higher education studies. Findings reveal that AI presents significant opportunities for enhancing historical inquiry, facilitating access to diverse historical sources, supporting personalized learning experiences, improving student engagement, and promoting innovative pedagogical approaches. At the same time, the proliferation of AI-generated content raises substantial concerns regarding source authenticity, algorithmic bias, assessment reliability, intellectual autonomy, data privacy, ethical accountability, and the potential weakening of students' critical historical reasoning and interpretive skills. The review further demonstrates that the integration of AI into history education extends beyond technological adoption to encompass broader questions of epistemology, knowledge production, disciplinary identity, and institutional governance. Consequently, African universities must develop context-sensitive and ethically grounded frameworks that balance technological innovation with academic integrity, responsible scholarship, and educational sustainability. The study proposes a human-centred framework for AI integration in historical pedagogy that emphasizes ethical governance, critical digital literacy, transparent assessment practices, and institutional accountability. By situating AI within the specific realities of African higher education and history education, this paper contributes to emerging debates on the future of humanities pedagogy in the digital age and provides policy, pedagogical, and institutional recommendations for the responsible and sustainable adoption of AI in historical scholarship and teaching.
Examining opportunities, challenges, and transformative learning
Rajni Bala1; Parul Kulshrestha1
The rapid advancement of Generative Artificial Intelligence (GenAI) has significantly transformed the higher education landscape, creating new opportunities for enhancing teaching and learning while simultaneously raising important ethical and academic concerns. As students increasingly integrate GenAI tools into their academic activities, understanding the factors influencing their adoption and the resulting educational outcomes has become a critical area of inquiry. This study examines students' adoption of Generative AI in higher education by exploring the influence of perceived opportunities and perceived challenges on adoption behavior and its subsequent impact on transformative learning. Drawing upon the Technology Acceptance Model and Transformative Learning Theory, the study conceptualizes perceived opportunities through dimensions such as improved productivity, personalized learning, and faster information access, while perceived challenges encompass ethical concerns, accuracy concerns, and academic integrity issues. A quantitative research design is proposed, targeting university students who actively engage with Generative AI tools for academic purposes. Data will be collected through a structured questionnaire and analyzed using Structural Equation Modeling (SEM) to test the proposed relationships. The study posits that perceived opportunities positively influence students' adoption of Generative AI, whereas perceived challenges negatively affect adoption intentions. Furthermore, the adoption of Generative AI is expected to contribute positively to transformative learning by fostering critical reflection, enhanced knowledge acquisition, and adaptive learning experiences. The findings are anticipated to provide valuable insights for educators, policymakers, and higher education institutions seeking to promote the responsible and effective integration of Generative AI in academic environments. By addressing both the benefits and challenges associated with GenAI adoption, the study contributes to the emerging discourse on technology-enabled educational transformation and responsible AI implementation in higher education.
Opportunities and ethical challenges
Tashi Wangmo A1
Artificial intelligence (AI) is increasingly entering classrooms worldwide, reshaping how high school students learn, interact, and prepare for future careers. This study investigates the role of AI in Bhutanese secondary education, focusing on its potential to personalize learning, support teachers in assessment, and enhance student creativity through digital tools. Using surveys, interviews, and classroom observations, the research highlights how students and teachers perceive AI applications in mathematics practice, language translation, and interactive learning environments. In addition to documenting current practices, the study explores how AI can reduce repetitive tasks for teachers, allowing them to devote more time to mentoring and holistic student development. It also examines the ways AI-driven platforms can adapt lessons to individual learning speeds, thereby addressing diverse student needs. The methodology emphasizes triangulation of data sources to ensure reliability, combining quantitative survey results with qualitative insights from classroom interactions. At the same time, the study critically examines challenges such as unequal access to technology, risks of dependency on automated systems, and ethical concerns around data privacy and adolescent well-being. Grounded in Bhutan's philosophy of Gross National Happiness (GNH), the paper argues that AI integration in high schools must balance innovation with inclusivity, cultural identity, and holistic development. The findings contribute to international debates on how secondary education can responsibly adopt AI to empower students while safeguarding ethical and social responsibility, offering recommendations for policymakers, educators, and researchers seeking to align technological progress with human values.
Taka Ramneesh1; Needesh Ramphul2; Jeetendre Narsoo3; Ajit Kumar Gopee3; Anjdilee Taka4
The rapid spread of Artificial Intelligence (AI), particularly generative AI technologies, is basically reshaping higher education systems globally. In Mauritius, where digital transformation and knowledge based economic development constitute strategic national priorities, the enhanced adoption of AI tools by tertiary students presents both transformative opportunities and complex societal risks. While AI has the potential to enhance personalised learning, academic efficiency, and research productivity, concerns persist regarding its effects on deep learning, critical thinking, academic integrity, and digital inequality. Despite the growing prevalence of AI usage among students, there remains no robust empirical evidence establishing its causal impact on academic performance within the Mauritian context or, more broadly, within Small Island Developing States. This study investigates the causal relationship between AI usage intensity and academic performance outcomes. Using a mixed-methods design, the research integrates econometric modelling with qualitative inquiry to capture both performance effects and behavioural dynamics associated with AI adoption. The study seeks to determine whether AI contributes to substantive learning gains or primarily enhances short-term performance metrics. The findings will inform higher education policy and institutional strategy in alignment with Sustainable Development Goals on quality education, innovation, and reduced inequalities. Eventually, the study aims to ensure that AI integration strengthens educational quality, promotes equitable access, and supports sustainable human capital development in Mauritius.
Hakem Hiba1
Student-centered learning (SCL) has long been promoted as a transformational pedagogical approach that has repositioned teaching and learning toward learner agency over teacher control. Yet despite decades of advocacy, its implementation remains uneven, context-dependent and frequently misunderstood. The rapid rise of artificial intelligence (AI) in education has brought these tensions into sharper focus, raising a critical question: what hidden inconsistencies in SCL is AI bringing into light? Drawing on a narrative review of literature published between 2015 and 2025, this presentation treats AI not merely as a tool that shapes learning, but as a diagnostic lens through which pre-existing contradictions within SCL become impossible to ignore; namely, whether learner autonomy, critical thinking, and self-regulation are genuinely cultivated or simply assumed in student-centered environments. What the literature reveals is that SCL has long been more idealized than practiced. Rather than introducing new problems, AI appears to have surfaced the gap between SCL's stated principles and its enacted realities. This presentation invite educators and researchers to move beyond idealized conceptions of SCL and engage critically with the evolving conditions that continue to shape and constrain its practice.
Emotionally intelligent gamification to boost engagement
Hezaimia Ouissal1; Fetsi Esma2
Can educational games respond to learners' emotions to improve engagement ? Emotionally intelligent gamification offers this possibility by integrating learners' emotional states into game-based learning. Unlike traditional gamification, which relies on fixed rewards such as points or badges, emotionally aware systems adjust feedback, challenge levels, and activities in real time based on cues like facial expressions, voice tone, or interaction patterns. This study examines the potential of emotionally intelligent gamification in EFL higher education classrooms, focusing on two groups of first-year students at the Ecole Normale Supérieure Al-Katiba Assia Djebar, Constantine. A questionnaire was used to explore learners' perceptions of engagement, motivation, and learning experience. Results suggest that emotionally adaptive gamification increases engagement, enhances motivation, and offers a sense of personalized learning. However, learners also raise concerns about privacy, ethical use of emotional data, and the need for teacher guidance. Overall, emotionally intelligent gamification can enrich learning when applied responsibly.
Understanding the optimal level of student engagement with artificial intelligence
Nada Baouta1
In the era where higher education is constantly evolving and undergoing rapid digital transformation, the integration of artificial intelligence (AI) raises central questions regarding its impact on cognitive engagement and the sustainability of learning outcomes. While, empirically speaking, moderate AI use may scaffold reasoning and enhance knowledge construction, excessive reliance still risks cognitive offloading and diminished critical thinking. This study, therefore, investigates the relationship between the intensity of AI use and self-reported critical thinking engagement while positioning critical thinking as a sustainability indicator of educational quality among undergraduate students. To this end, a semi-structured survey was administered to 60 students at École Normale Supérieure El Katiba Assia Djebar Constantine (ENSC) in Algeria. The survey measured AI frequency of use, AI dependency, and critical evaluation engagement via validated Likert-scale items. Moreover, the respondents were divided into low, moderate and high AI users and analyzed using ANOVA to examine the inverted U-shaped relationship between AI engagement and self-reported critical thinking. Preliminary findings show that moderate AI users reported the highest levels of critical thinking engagement, whereas both low and high users demonstrated comparatively lower engagement. Such findings corroborate Sweller's cognitive offloading theory and align with previous observations, that technology-mediated scaffolding optimizes cognitive involvement only within an optimal zone. This evidence suggests that sustainability in AI integration in higher education context requires balanced implementation strategies that encourage moderate engagement, preserving critical thinking while leveraging technological affordances, with implications for curriculum design, pedagogical policy, and institutional guidelines in digitally transforming universities.
Sally Abdulbast Jasem1
The integration of Virtual Environments for Training and Simulation (VETS) has basically reinterpreted the experimental process of education by reconciling the theory with practice through high accuracy and risk-free frameworks. This abstract investigates how VETS utilizes the key foundations of protection, adaptability, and reliability to modify proficient instructions while methodically turning its intrinsic restrictions into pedagogical privilege. Basically, VETS employs Experiential Learning Theory, giving the chance to learners to take part in the process of learning by working within frameworks that are either unsafe or logistically unachievable to be reproduced in the real world. In medicine, this demonstrates virtual anatomy labs which banish the necessity of cadavers; in aviation and engineering, it presents a platform for stress-testing both machinery and making human decisions under "black swan" circumstances. In addition, the emergence of AI-powered NPCs and haptic feedback has established a level of intellectual and tangible accuracy that promotes authentic muscle memory and social existence. Nevertheless, the acquisition of VETS is not free from obstacles. Extreme application has bad sides, the cyber-sickness of encompassing boundaries, and the risk of over- confidence in neutralized virtual frameworks introduce crucial challenges. Current pedagogical pattern tackles these by changing restriction into advantages. High expenses are reduced through cloud-based consortiums that equalize access. Physical unease is dealt with through micro-learning modules which intensify attention, moreover, the consistency of models is respond to by introducing unsystematic factors to create actual-world flexibility. Eventually, VETS is changing from an additional tool to a basic curriculum need. By making use of biometric analytics to present high-personalized estimation, these frameworks provide accuracy in the field of education never attained before. Through 2026, the concentration is still on closing the "Digital Divide" to guarantee these benefits are approachable within all socioeconomic levels, proving that the future expectations of training is to be comprehensive and unconventional.
Exploring the impact of generative AI on independent learning in higher education.
Akal Chaimaa1; Abdelouahid BENADDA2
The growing adoption of generative Artificial Intelligence (AI) tools in higher education is transforming the ways students' access information, complete assignments, and engages with learning. While applications such as ChatGPT and other AI-powered platforms offer immediate support for writing, researches, and problem-solving tasks, offering new opportunities for personalized and efficient learning, concerns about the growing reliance on these technologies have emerged regarding students' intellectual autonomy, critical thinking abilities, and long-term academic development. This study examines the extent to which generative AI functions as a valuable educational aid while also contributing to academic dependency among university students. A mixed-methods approach was adopted to explore students' perceptions and practices regarding AI use in academic contexts. A questionnaire was administered to 30 undergraduate students from different academic disciplines at a public university, complemented by ten semi-structured interviews. The findings indicate that most participants viewed AI as a useful learning aid that enhances productivity and facilitates access to information. However, many respondents expressed concerns that excessive reliance on AI may reduce independent research, weaken critical evaluation skills, and decrease engagement with course materials. Interview data further revealed that students generally recognize the educational value of AI but remain uncertain about the boundaries between legitimate academic assistance and overreliance. The study concludes that the educational impact of generative AI depends largely on how these technologies are integrated into teaching and assessment practices. It recommends promoting AI literacy and responsible use to ensure that AI supports, rather than replaces, students' critical thinking, creativity, and independent learning.
Theoretical Foundations, Analytical Approaches, and Applications
Bashar Abdulkareem Alali
This article offers a critical narrative review of multimodality as a theoretical and analytical paradigm within English for Specific Purposes (ESP). Drawing on social semiotic theory, systemic functional multimodal discourse analysis, and multimodal interactional analysis, it traces the intellectual history of multimodal inquiry and maps its main theoretical frameworks, methodological approaches, and applied domains. The review argues that professional communication is multimodal by nature: meaning is spread across verbal, visual, spatial, and gestural resources, and ESP research that sidelines this complexity ends up with an incomplete picture of specialised communicative practice. The article looks at how multimodal and genre-analytic frameworks have come together, at the digital transformation of professional semiotic landscapes, and at what this all means for ESP curricula. It closes by identifying four directions the field is moving in, including the implications of artificial intelligence, computational multimodal analysis, data visualisation, and cross-cultural perspectives.
Students’ Perspectives on the Use of NotebookLM in Nursing Education
Angelina Kljajić1; Nevena Živković1
The use of information and communication technologies (ICT) and digital tools in the teaching process has become an imperative of contemporary education. This paper presents a study conducted by two practising teachers. Its aim is to examine, understand, and describe, within the theoretical framework of educational sciences, secondary school students’ experiences of using the digital tool NotebookLM in nursing education. For the purposes of the study, a questionnaire was developed and data were collected through an anonymous student survey. The study involved 47 second-year students, 39 females and eight males. The findings indicate that the use of NotebookLM in teaching is both justified and necessary, not only from the perspective of educational sciences but also from that of the students themselves. According to the respondents, the tool helps them connect theoretical and practical knowledge more easily, reduces anxiety associated with assessment, encourages independent exploration of additional sources of knowledge, and supports an individualised approach to learning by allowing students to learn at their own pace. Students also consider the use of digital technologies in medical schools important. They report that such technologies support long-term retention of content, facilitate mastery of the subject matter, and significantly enhance the overall learning process.
Imen Talbi1; Lilia Smida2; Jamel-Eddine Gharbi2
The issue addressed in this research is, on the one hand, the impact of the use of Artificial Intelligence (AI) tools on learner engagement. Absenteeism, plagiarism, outsourcing of activities and demotivation of students are experienced consequences of the development of AI platforms. On the other hand, the use of AI tools seems to lead to a weakening of learners' creativity. Based on the theory of hybrid human-AI intelligence, we try to test the relationship between the use of AI tools, learner engagement and creativity. We conducted a survey of a sample of students. We used scales validated in the literature. The results reveal a significant positive impact of use of AI tools on learning engagement and creativity. Similarly, this engagement has a positive influence on creativity. Indeed, the results show that engagement plays the role of a total mediation between the use of AI tools and the creativity of learners. The practical importance of these findings lies in ensuring that the integration of AI into higher education strengthens learners' engagement and develops their creativity.
A driver of engagement and performance?
Ghofrane Barhoumi1; Jamel Eddine Gharbi1
In the digital age, student engagement in higher education remains a major challenge, despite the proliferation of technological tools. Artificial intelligence (AI), and in particular chatbots like ChatGPT, represents a unique opportunity to boost engagement and improve academic performance, while offering new perspectives in educational marketing. This study analyzes how the perceived attributes of AI (usefulness, ease of use, quality of interaction, personalization), combined with motivational and social factors, influence the engagement and, ultimately, the perceived academic performance of Tunisian students. Based on an online questionnaire of 200 students and a structural equation analysis (SEM), the results confirm that student engagement acts as a central mediator, amplifying the effect of technological, motivational and social determinants on academic success. These findings show that the strategic integration of AI in education, perceived as useful and interactive, can transform the learning experience, particularly in emerging contexts with limited resources. For educational marketing, these findings highlight the importance of designing student-centric digital experiences to maximize engagement and performance.
Imen Talbi1; Jamel-Eddine Gharbi2
The issue addressed in this research is, on the one hand, the impact of the use of Artificial Intelligence (AI) tools on learner engagement. Absenteeism, plagiarism, outsourcing of activities and demotivation of students are experienced consequences of the development of AI platforms. On the other hand, the use of AI tools seems to lead to a weakening of learners' creativity. Based on the theory of hybrid human-AI intelligence, we try to test the relationship between the use of AI tools, learner engagement and creativity. We conducted a survey of a sample of students. We used scales validated in the literature. The results reveal a significant positive impact of the dimensions of the use of AI tools on the three dimensions of learning engagement and four dimensions creativity. Similarly, this engagement has a positive influence on creativity. Indeed, the results show that engagement plays the role of a total mediation between the use of AI tools and the creativity of learners. The practical importance of these findings lies in ensuring that the integration of AI into higher education strengthens learners' engagement and develops their creativity.
Imen Talbi1; Jamel-Eddine Gharbi1
Artificial intelligence (AI) is transforming learning practices in higher education and is attracting growing interest in its impact on student creativity. This research analyzes the effect of the use of AI tools on creativity. She wishes to verify the moderating impact of AI capabilities and skills on the relationship between the use of AI tools and creativity. The links we want to test are based on the theory of capabilities, the componential theory of creativity, UTAUT2 and the theory of human-AI complementarity. Data collection was carried out by a survey of a sample of students. The measurement scales come from the literature. The results show that the use of AI tools positively influences students' creativity. Similarly, the results confirm the moderating and intensifying role of AI capabilities and skills on the relationship between use and creativity. Thus, this research highlights the responsibility of the university institution to develop the ten forms of general capabilities and the skills of students in AI so that the use of AI means leads to the development of learners' creativity.
Transforming humanistic education with artificial intelligence.
Prakash Meena1; Vikas Rana2; Pramendra Kumar Meena3
Artificial intelligence's (AI) quick development has started to change education, especially in the historically human-centered fields of the social sciences and humanities. This chapter, "Evolving Beyond Chalkboards to AI Chatbots: Transforming Humanistic Education with Artificial Intelligence," examines how intelligent tutoring systems and chatbots in particular are revolutionizing academic inquiry, critical thinking, pedagogy, and learner engagement in humanistic education. Humanistic disciplines, which have historically been based on interpersonal communication, introspection, and contextual interpretation, are increasingly interacting with data-driven customization, machine learning, and natural language processing. This chapter explores the potential and drawbacks of using AI technology into curriculum design, classroom interactions, and evaluation procedures. It critically examines how AI might improve digital literacy and critical inquiry—skills essential to humanities education—create individualized learning pathways, and mimic Socratic debates. The chapter also addresses the practical, intellectual, and ethical difficulties that surround AI in the humanities, including as prejudice, dehumanization, maintaining academic integrity, and maintaining empathy and cultural sensitivity. This chapter emphasizes the indispensable value of human insight, creativity, and moral judgment in a technologically advanced academic setting while offering a thorough overview of AI's impact on humanistic education through case studies, practical classroom applications, and forward-looking frameworks. Objectives 1. Investigate the use of AI tools—particularly chatbots—in humanistic education. 2. To investigate how AI affects teaching, critical thinking, and student involvement. 3. To showcase case studies and real-world AI applications in humanities education.
Culture, learning and social relations
Edmont Pasipamire1
The rapid development of generative artificial intelligence (GenAI) is changing teaching practices, pedagogy approaches and the relationship in the classroom. While current scholarship is increasingly exploring the use of artificial intelligence in teaching, learning, assessment and academic support, its wider impact on culture, human interaction and social relations in education still needs to be understood. This paper explores how GenAI is reshaping education and learning, with a particular focus on the interface between technological change, cultural contexts and social relations. The study adopts a literature review methodology, drawing on the academic literature on general artificial intelligence, artificial intelligence in education, teaching, learning, learning assessment, digital literacy, academic integrity, learning culture and teacher-student relations. The relevant literature is critically reviewed and analysed thematically to identify recurring perspectives, new discussions, opportunities, challenges and gaps in the existing scholarship. In particular, the review examines how AI-enabled learning affects teaching roles, learning pathways, student involvement, evaluation, academic integrity, human agency and teacher-student relations. The review shows that GenAI does not only introduce new technological tools into education, but also contributes to broader changes in pedagogical practice, institutional culture and pedagogical relations. The literature highlights personalised learning, reinforced pedagogical support, creativity and access to knowledge, but also raises concerns about academic integrity, over-reliance on artificial intelligence, digital inequalities, cultural differences, trust, authorship and the changing role of pedagogy. The document argues that understanding Genoa in education requires not only focusing on technical competences but also on the cultural and social dimensions of teaching and learning.
Artificial intelligence, quality education (SDG-4) & the axis of introspection
Mazhar Shamshi Ansary1; Ayush Mazumdar2; Santosh Behera3; Jayanti Das4
The potential of Artificial Intelligence (AI) in Education has emerged as the harbinger of radical transformation in the contemporary world and has also opened Pandora's box. The wider impact of AI can significantly contribute to promoting quality education as part of the United Nations Sustainable Development Goal (SDG) 4, which forms a crucial pillar of positive change. Quality education deals with fostering efficiency and creativity of learners; developing knowledge, enhancing social and interpersonal skills, problem-solving skills and developing global citizenship education and many more. Though AI has been in existence since the middle of the 20th century, it has been popularised rapidly and has made a rapid breakthrough over the last decade. The capability and praxis of Artificial Intelligence, complemented by its impressive features made researchers, policymakers, academicians take note of its transformative capabilities in the education sector with a possible synergy between the two. In the age of digitalisation, students around the globe want and are nudged by socio-cultural factors to embrace technological innovations. This led the Government of India to invest significantly in the AI ecosystem as part of its SOAR initiative, which aims to build AI awareness and skilling for class 6-12 students and educators. The Government of India, through a handholding and collaborative approach with the UGC, AICTE, SWATAM and various IITs, has been engaging in seminars, conferences, conclaves, workshops, etc. The paper delves into finding the contribution of Artificial Intelligence (AI) in disseminating quality education as mentioned in SDG 4, as envisaged by the Government and UNESCO. The political-economic concerns of authoritative allocation of resources lead us into a situation of introspection in this age where sustainable development must not compromise the needs of the present generation because failure to plan at this juncture would lead to a severely cripped future generation. Hence, we argue that the Madhyam Marg approach is needed in order to address the challenge emanating from the demographic burden vs demographic dividend discourse. The methodological tenet of this paper engages in a qualitative and interpretive analysis of relevant literature in the field of enquiry to address the issues raised, as well as engage with the theory of critical pedagogy, which raises the environmental aspect of Artificial Intelligence utility and the digital divide.
A Systematic Review of Impact and Challenges
Malathi Letchumanan
Artificial intelligence (AI) chatbots are increasingly being integrated into mathematics learning. Although previous reviews have examined the impact of chatbots on learning, they have not explicitly examined how mathematics students utilise chatbots to learn mathematics. To address this gap, following the PRISMA guideline, this study systematically reviewed 17 articles published between 2020 and 2026 to summarise the positive impacts and challenges of chatbot use for learning mathematics. The findings showed that chatbots provide a compelling platform to enhance students’ cognitive and affective learning outcomes. However, the findings also raised concerns about ethical issues, accuracy, reliability, and the suitability of using chatbots to solve mathematics problems. Lessons learned from this review provide recommendations to practitioners, policymakers, and chatbot developers.
Descriptive findings from a pilot survey in an Algerian STEM context
Bayane Guellal1
The use of Artificial intelligence (AI) has recently predominated different dimensions of higher education. In the context of English-medium instruction (EMI) specifically, both students and teachers seek AI to maneuver content and language simultaneously, especially in STEM settings that require both mastery of technical terminology and explanatory precision. Yet, research on how STEM faculty perceive AI tools in their teaching remains almost scarce in Algeria and the MENA region. Thereto, this study reports descriptive findings from 30 faculty members at Ferhat Abbas University's Faculty of Technological Sciences, who are all teaching technical courses in English, using the TAAI-EMI questionnaire developed and validated for EMI contexts in non-Anglophone settings. The study examined five dimensions of attitudes toward AI in EMI: perceived usefulness, perceived ease of use, behavioral intention, self-efficacy, and anxiety. The questionnaire included 29 Likert-scale items rated from 1 to 5. Consequently, the results revealed the following: perceived ease of use (M = 4.04, SD = 0.86) > behavioral intention (M = 3.93, SD = 0.79) > perceived usefulness (M = 3.58, SD = 0.90) > self-efficacy (M = 3.38, SD = 0.93) > anxiety (M = 2.70, SD = 0.82). Moreover, item-level results suggest that respondents perceived AI as relatively easy to use and expressed relatively positive intentions to continue using it, while their judgments of usefulness and especially their confidence in using it effectively for EMI were more moderate. Anxiety was comparatively lower overall, but concerns pertaining to privacy, security and concerns AI translation accuracy for technical EMI terminology remained notable. Because the study used a small convenience sample (N = 30), the findings are descriptive and exploratory and should not be generalized to all Algerian university teachers or all STEM/EMI teachers. Nevertheless, the study suggests that teacher support should move beyond basic tool introduction and more toward AI literacy, EMI-specific pedagogical integration, confidence-building, accuracy checking, and responsible use.
Barriers, motivations and perspectives for AI integration in France and Tunisia
Amina Ben Mabrouk1; Hayet Saoudi2; Safia El-Bok3
Environmental and sustainability education (EESD) is a stated educational priority in both France and Tunisia. In the digital age, the rapid emergence of artificial intelligence (AI) opens up unprecedented possibilities for teaching EESD, but also raises uncertainties and ethical concerns. This contribution draws on an original quantitative survey conducted among secondary school teachers in France and Tunisia. The study examines teachers' actual EESD practices, their representations of digital tools and artificial intelligence, as well as the obstacles and levers that shape their relationship with the integration of these technologies. Obstacles include insufficient training, inadequate pedagogical resources, and overloaded curricula. The levers identified include personal commitment, awareness of climate issues, and interest in pedagogical innovation. The initial results reveal significant contrasts between the two countries. In France, teachers demonstrate a greater familiarity with digital tools, accompanied by a weariness with recurrent institutional pressures. In Tunisia, by contrast, a strong enthusiasm for EESD and AI is hindered by material constraints and limited access to available resources. Nevertheless, similarities emerge: teachers in both countries express a shared sense of a critical lack of targeted training on AI applied to EESD, as well as a fear that AI may exacerbate pre-existing inequalities between schools. Finally, the contribution offers concrete perspectives for a responsible, ethical and context-sensitive integration of AI into EESD, in line with the social responsibility of educational actors.
Implications for TeachersinCharge in early childhood education
Akiva Onika Sarita La Geer-Jeremiah1
Artificial intelligence (AI) is increasingly influencing leadership and pedagogical practices in early childhood education (ECE), creating both opportunities and challenges for Teachers in Charge (TiCs). This article examines the implications of AI integration for leadership, pedagogy, and ethical governance in preschool settings. Drawing on a systematic review of peer-reviewed literature published between 2020 and 2026, the study synthesizes evidence through the lenses of Sociocultural Learning Theory, Transformational Leadership Theory, and Ecological Systems Theory. The findings suggest that AI can enhance personalized learning, language development, adaptive instruction, and teacher professional learning when implemented within child-centred and relational pedagogical frameworks. However, concerns related to data privacy, algorithmic bias, equity, developmental appropriateness, and ethical governance remain significant barriers to responsible adoption. The review highlights the critical role of TiCs in fostering AI literacy, promoting pedagogical resilience, supporting ethical decision-making, and ensuring that technological innovation aligns with the developmental needs and rights of young children. The article argues that effective AI integration requires adaptive, collaborative, and ethically informed leadership approaches. Implications for policy, professional development, and future research are discussed, with particular attention to strengthening responsible AI use in ECE contexts.
Preliminary findings and challenges
Tatjana Jovcheska1
The rapid integration of digital technologies in education has redefined learning environments, requiring teachers to examine and adapt their teaching practices. This preliminary study, conducted as part of a doctoral research project, investigates ELT teachers' understanding of the digital pedagogy affordances, their current practices, and the challenges they encounter with their implementation in the language classrooms in North Macedonia. Current digital pedagogy frameworks emphasise the potential of digital technologies to support active learning, collaboration, differentiated learning, and multimodal meaning making. However, research indicates that many teachers have limited awareness of digital concepts and frameworks. Accordingly, this study looks into teachers' understanding of digital pedagogy, their intentional and/or unintentional application of its affordances, their perceived challenges, and their willingness to pursue further professional development. This small - scale study employed a quantitative research design to analyse ELT teachers' perspectives and experiences with digital pedagogy. The preliminary findings show that even though teachers demonstrate limited awareness of digital pedagogy affordances, many already integrate elements of these affordances into their teaching, often implicitly. Moreover, teachers express a strong willingness to extend their knowledge and develop their digital pedagogical competence. These findings underscore the need for professional development opportunities that help teachers move beyond the use of technology towards meaningful ELT practices.
The mediating role of organizational innovation
Sondeep Kaur1
Artificial intelligence (AI) is increasingly transforming higher education institutions by enhancing administrative efficiency, academic decision-making, and stakeholder engagement. Despite growing investment in AI technologies, limited research has examined how institutional AI readiness contributes to sustainability performance. This study investigates the relationship between AI readiness and sustainability performance in higher education institutions and examines the mediating role of organizational innovation. A quantitative research design was adopted using a cross-sectional survey of academic and administrative personnel employed in public and private higher education institutions (N = 500, synthetic demonstration dataset). Data were collected using a structured questionnaire that measured AI readiness, organizational innovation, and sustainability performance. AI readiness was assessed across technological infrastructure, human competencies, leadership support, and organizational preparedness, while sustainability performance was evaluated across economic, social, and environmental dimensions. The proposed research model was tested using Structural Equation Modeling (SEM) with maximum likelihood estimation and bootstrapping (5,000 resamples). The findings reveal that AI readiness has a significant positive direct effect on sustainability performance (β = 0.546, p < 0.001) and a significant positive effect on organizational innovation (β = 0.647, p < 0.001). Organizational innovation, in turn, significantly enhances sustainability performance (β = 0.314, p < 0.001) and partially mediates the relationship between AI readiness and sustainability performance (indirect effect = 0.202, 95% CI [0.141, 0.263]). These results suggest that innovative capability is a critical mechanism through which AI readiness translates into sustainable institutional outcomes. The study contributes to the emerging literature on AI-enabled transformation in higher education by providing empirical evidence of the strategic role of organizational innovation in achieving sustainability objectives. The findings offer practical implications for university leaders and policymakers seeking to leverage AI technologies to promote responsible innovation, institutional resilience, and long-term sustainable development.
Curriculum structures and trends of Singapore polytechnic diploma programmes
Liu Chaofan1
Artificial intelligence (AI) education is expanding rapidly across higher education, but existing research has focused predominantly on universities, AI literacy frameworks, and pedagogical innovation, with comparatively limited attention to curriculum design in practice-oriented higher education. Addressing this gap, this study examines Applied Artificial Intelligence diploma programmes across Singapore's five polytechnics through a qualitative, document-based comparative analysis of publicly available institutional materials, including programme descriptions, curriculum structures, module information, and learning outcomes. The analysis explores programme structures, emerging curriculum trends, and practice-oriented educational characteristics. The findings reveal substantial curricular convergence across institutions, with programmes consistently organized around programming, data analytics, machine learning, software development, and project-based learning, while also demonstrating meaningful institutional differentiation in technological emphasis, ethical orientation, industry integration, and interdisciplinary application. Some programmes foreground emerging technical ecosystems such as Generative AI, MLOps, intelligent automation, and deployment environments, whereas others place stronger emphasis on responsible AI, ethics, industry certifications, or interdisciplinary digital integration. The analysis further shows that experiential learning, workplace engagement, applied projects, internships, and industry responsiveness constitute core features of Applied AI education within Singapore polytechnics. The study argues that Applied AI curriculum development in practice-oriented higher education extends beyond technical training alone and increasingly reflects a hybrid curriculum model combining technical competence, applied learning, industry alignment, and responsible AI awareness. By focusing on Singapore's polytechnic sector, this study contributes to emerging scholarship on AI education by highlighting how curriculum architecture operationalizes AI education within diploma-level, non-university contexts.
Perspectives from Nigeria and Tanzania
Cynthia Adaeze Onuegwunwoke1; Joseph Paschal Mahona2
Artificial intelligence (AI) is rapidly reshaping teacher education, yet the perspectives of English language teacher educators in Sub-Saharan Africa had remained underexamined. This study investigated how teacher educators in Nigeria and Tanzania perceived, accepted, and integrated AI tools into their pedagogical practice. This was anchored on the Technology Acceptance Model (Davis, 1989) and the Technological Pedagogical Content Knowledge framework (Mishra & Koehler, 2006). A comparative descriptive survey was conducted with 120 educators (60 from Nigeria, 60 from Tanzania) using a six-subscale questionnaire (alpha = .76-.89) analyzed descriptively and inferentially. Teacher educators in both countries reported moderate to high perceived usefulness (Nigeria M = 3.73; Tanzania M = 3.60) and moderate perceived ease of use (Nigeria M = 3.05; Tanzania M = 3.22), with no significant cross-national difference. Both samples displayed a consistent TPACK gradient, with strong Content and Pedagogical Knowledge but weaker Technological Knowledge and integrated TPACK for AI. Tanzanian educators reported higher institutional support than their Nigerian counterparts (t(118) = -2.29, p = .024, d = -0.42), while ethical concerns around academic integrity and algorithmic bias against local languages were uniformly high in both countries. Lack of training was the most frequently cited barrier to adoption in Nigeria (73.3%) and Tanzania (68.3%). Regression analysis showed that perceived usefulness, perceived ease of use, TPACK, and institutional support each predicted intention to use AI (R2 = .48), and institutional support moderated the relationship between perceived usefulness and intention. These findings inform teacher education policy, curriculum reform, and professional development across Nigeria, Tanzania, and African contexts.
Nourchen Bouzidi1; Jamel-Eddine Gharbi2
This research attempts to identify the factors that explain the mismatch between university education and market needs in Tunisia. With reference to morphogenetic theory, we empirically verified the direct and interactive influence of the structure, culture and university agency on the adequacy between training and market needs. A survey was conducted with 240 respondents (120 students and 120 teachers). The results confirm the positive impact of structure and culture on the adequacy of university education and market needs. The agency's impact on students has been validated. On the other hand, agency among teachers does not have a significant effect. This research empirically verifies the morphogenetic theory. It shows that higher education in Tunisia is in morphostasis. It is in a structural and cultural determinism. Social interactions in academia are not yet capable of providing actors with the agency power and status to transform the dominant traditional structure and culture.
A TPACK-AI approach to transformative teacher professional development
Fouad Boulkroun1
The rapid integration of artificial intelligence (AI) into educational ecosystems has fundamentally disrupted the competency expectations placed upon classroom teachers, necessitating a critical re-examination of teacher professional development (TPD) frameworks. While the Technological Pedagogical Content Knowledge (TPACK) framework has long guided educators in navigating the intersection of technology, pedagogy, and content knowledge, its foundational architecture was not designed to accommodate the systemic, ethical, and epistemological dimensions that AI introduces into contemporary teaching and learning environments. This paper draws on the emerging TPACK-AI construct, which extends the foundational TPACK model by embedding AI-specific literacy competencies – including algorithmic awareness, data ethics, and human-AI collaboration – as distinct knowledge domains within a reconceptualised professional knowledge architecture. Employing a systematic conceptual review, this study examines how TPACK-AI reorients TPD from tool-centric technology integration toward deeper, sustained pedagogical transformation, addressing persistent gaps in existing literature including the absence of standardised AI competency benchmarks, inadequate formative feedback mechanisms, and inequitable program access. A central contribution is the development of a TPACK-AI TPD Checklist, operationalising the framework across six interrelated domains: foundational AI knowledge, algorithmic awareness and data ethics, pedagogical AI integration, human-AI collaboration, institutional alignment, and equity-responsive practice. The findings suggest that repositioning TPD within a TPACK-AI framework, structured through a domain-specific checklist, enables coherent, competency-driven professional learning pathways responsive to the ethical and contextual complexities of AI in education. This paper offers a theoretically grounded and practically actionable contribution for teacher educators, curriculum designers, and institutional leaders seeking transformative, equity-conscious outcomes in AI-integrated professional learning contexts.
Navigating AI transformation in Global higher education
Janvi Anand Rathi1; Shailendra Naresh Kulkarni1; Mensur Zeqiri1
Artificial intelligence (AI) is rapidly transforming higher education by reshaping teaching, assessment, research, governance and institutional decision-making. While generative AI offers opportunities for personalised learning, administrative efficiency and research support, its adoption also raises concerns regarding academic integrity, data governance, ethical oversight, equity, cultural representation and public trust. This conceptual and policy-oriented paper examines how higher education institutions can move beyond fragmented or reactive responses towards strategically governed, ethically responsible and socially inclusive AI implementation. Drawing on interdisciplinary literature from higher education policy, AI ethics, digital governance and critical knowledge-system perspectives, the paper introduces the Institutional Readiness and Ethical Integration (IREI) Model. The model assesses institutional preparedness across four dimensions: strategic readiness, operational readiness, ethical readiness and evaluative readiness. The paper identifies three institutional approaches to AI adoption: restrictive, experimental and strategically governed. It argues that sustainable AI integration requires transparent governance, human oversight, faculty and student capacity building, assessment integrity frameworks and culturally responsive policies. Particular attention is given to the risk that dominant AI systems may reproduce linguistic, cultural and epistemic inequalities, especially for institutions and learners in underrepresented regions of the Global South. The paper concludes that AI in higher education should be governed as a human-centred public good that advances academic quality, institutional trust, ethical accountability and inclusive global development.
A critical examination of contemporary template distortions
Jacob Filgona1; John Sakiyo2
Lesson planning is a central practice in teacher education and instructional design, yet contemporary lesson plan templates have undergone extensive modification without corresponding theoretical justification. Many templates now exhibit conceptual inconsistencies that distort the pedagogical purposes of their constituent components. This article introduces the concept of functional integrity—the principle that each lesson plan component should perform a distinct instructional function without duplication, substitution, or conceptual overlap with other components. It further proposes sequential coherence – the principle that lesson plan components should be ordered according to the temporal and logical dependencies of instruction. Drawing on instructional design traditions including behavioural objectives, instructional events, alignment theory, and teacher planning research, the article identifies three recurring threats to functional integrity: functional distortion (inappropriate information placed within a component), functional duplication (identical information repeated across components), and functional displacement (one component performing another's role). Through systematic analysis of common lesson plan components, the article develops a principle-based framework for evaluating template quality. The framework provides school administrators, teachers, teacher educators, and policymakers with diagnostic tools. In order to enable learner-centred pedagogy, formative assessment, and self-regulated learning, the study suggests that lesson planning quality should be evaluated based on the preservation of functional integrity and sequential coherence rather than template complexity or administrative completeness.
The role of organisational, social and policy innovations
Amira Fadhlaoui1; Hmaidi Khoubayeb2; Jamel-Eddine Gharbi1
The issue addressed in this study is the low level of collaboration between university research and teaching and the Tunisian business sector. The aim of the research is to test the influence of organisational, social and political innovation on university-industry collaboration. The data used to test our conceptual model are drawn from the Global Competitiveness Reports published by the World Economic Forum and the Global Innovation Reports published by WIPO. Regression analyses show that university-industry collaboration is significantly influenced by the three dimensions of innovation, namely organisational, social and political innovation.
Professional development challenges and education policy
Frida Alfred Mbwafu1
The integration of Artificial Intelligence (AI) is highly changing the way of teaching and learning everywhere around the world. In Tanzania, English teachers in lower public secondary schools are struggling to keep up with AI in their classrooms. However, many of them are not equipped with enough training to help them cope with the technological integrations. This study aimed to find out how ready the English teachers are for teaching with AI, and what challenges they face in terms of professional development and policy support. Using the TPACK framework, the main question was: Are Tanzanian public secondary school English teachers prepared for AI-powered classrooms, and what challenges stand in their way? Most research so far has focused on AI in universities and Western countries; there is little information about how prepared secondary school teachers in developing countries like Tanzania really are. To fill this gap, the study used interviews and group discussions with English teachers from three public schools. The results showed that teachers often do not know much about AI, have little access to technology, and rarely get in-service training. National policies on AI in schools are still vague. Many teachers are enthusiastic about AI's potential but feel unsure and unsupported as they try to make it work in their classrooms. The study suggests English teachers need ongoing training, better digital infrastructure, and inclusive AI policies. Ultimately, helping teachers feel confident and prepared is key to making AI-supported education fair and successful for every student in Tanzania.
New competencies for academic leaders
Sruthi Pillai1; Zunaith Ahmed2; R Ramakrishnan3
Artificial intelligence is transforming higher education with is growing use in institutional structures, decision-making strategies, academic work, and leadership roles. As institutions negotiate hitherto unheard-of technical change, Academic leaders are challenged to balance between creativity, moral responsibility, organizational resiliency, and stakeholder expectations Though there is growing interest in artificial intelligence in education, little study on the leadership skills required to guide institutions through AI-driven transformation has been done. This conceptual article develops an AI Leadership Competency Framework for institutions of higher learning. To enable academic leaders to gain the core competencies required to operate AI-enabled businesses, the study combines strategic leadership theory with digital transformation literature, change management, and responsible AI governance. The proposed framework consists of five connected skills: Digital literacy, Adaptive change leadership, Human-centered stakeholder engagement, Strategic AI vision, and Ethical governance. Long-term corporate results as well as institutional preparation for artificial intelligence implementation depend on these skills. Apart from technological knowledge, the study shows artificial intelligence leadership as a complex ability comprising moral judgment, organizational learning, trust-building, and strategic foresight. Good artificial intelligence leadership also helps a company be more flexible, creative, accepting of its staff, and able to provide better education. It also lowers hazards linked with administrative mistakes, transparency problems, and algorithmic prejudice. By offering a comprehensive leadership framework for AI-enabled companies, this study supports scholarship in management and higher education. It provides helpful guidance for college leaders, legislators, and educational leaders working to create leadership capacity in reaction to fresh technological disruptions. The framework also lays the groundwork for later empirical studies on the connection between AI leadership skills and institutional transformation results in higher education.
A management framework for governance, trust, and organizational transformation
Zunaith Ahmed1; Sruthi Pillai2; R Ramakrishnan3
Through uses in instruction, learning, assessment, administration, and institutional decision-making, artificial intelligence (AI) is revolutionizing higher education. AI presents great possibilities for personalization, efficiency, and innovation. But its use also sparks issues about openness, trust, governance, academic integrity, accountability, and organizational readiness. Most research concentrates on technical execution or pedagogical ramifications; little focus is paid to the organizational and management aspects of responsible artificial intelligence acceptance. This article presents a conceptual management framework for ethical artificial intelligence use in colleges and universities. Drawing from theories of organizational transformation, technology acceptance points of view, institutional governance data, and ethical innovation concepts, the framework pinpoints the key organizational elements influencing effective artificial intelligence integration. More precisely, the study contends that AI governance frameworks, moral standards, leadership dedication, stakeholder involvement, and institutional AI knowledge taken together determine how much a company trusts AI systems. Institutional trust mediates the relationship between these organizational conditions and responsible AI adoption, which in turn contributes to organizational transformation. The recommended model views are responsible for artificial intelligence use as a strategic organizational competence rather than mostly a technical endeavor. It stresses how universities could balance responsibility with invention while still ensuring compliance with society's expectations and educational standards. The study supports management scholarships by using responsible innovation theory in the context of higher education and offering leaders a working governance framework for institutions. The framework could enable colleges, legislators, and educational administrators to develop sustainable AI adoption strategies that enhance organizational performance while maintaining ethical standards, stakeholder trust, and educational quality in the age of artificial intelligence.
Opportunities and challenges
Nikos Tsipas1; Lamprini Seremeti1
Emerging technologies have radically transformed the modern educational ecosystem. Among these technological innovations, Business Intelligence (Business Intelligence, BI) is emerging as an increasingly important driver of educational transformation, supporting stakeholders in decision-making processes related to teaching, planning, quality assurance policies, and the administration of educational institutions. This study examines the role of BI in education through a synthesis of the existing literature. It focuses on the main arguments for and against the use of BI in educational settings, highlighting both its potential benefits and limitations. On the one hand, from a theoretical perspective, Business Intelligence is expected to support evidence-based decision-making, improve the monitoring of student performance, and enhance the effectiveness of educational institutions. On the other hand, in practice, its implementation in the educational routine raises concerns regarding data protection issues, the assessment of data quality, and the suitability of quantitative indicators for measuring complex educational effectiveness. By combining theoretical expectations with practical challenges, this study aims to provide a balanced understanding of how Business Intelligence can contribute to education.
Exploring generative AI for lesson planning and instructional innovation in EFL contexts
Djaber Oumaima1; Ghaouar Nesrine1
Generative artificial intelligence (GenAI) is increasingly being used as a support tool in educational practices, including lesson planning. This study aims to explore EFL teachers' perceptions of the use of generative artificial intelligence (GenAI) for lesson planning. To achieve this aim, a questionnaire was distributed to 25 EFL teachers at Badji Mokhtar Annaba University, Algeria, to examine their use of GenAI, perceived benefits, challenges, and recommendations for its effective and responsible integration into lesson planning. The findings show that GenAI was widely used by the participants, particularly for designing classroom activities, preparing teaching materials, and developing quizzes and assessments. Teachers generally reported positive perceptions, especially regarding time saving, faster idea generation, and the development of engaging activities. However, they also identified concerns related to accuracy and reliability, mismatch with students' proficiency levels, and generic or repetitive outputs. Teachers therefore recommended verifying AI-generated content, maintaining professional judgement, using AI as a supplementary tool, and developing training and prompting skills. These findings suggest that GenAI can support lesson planning effectively when combined with teachers' pedagogical knowledge, critical evaluation, and professional judgement.
Mandu S. Umoren1; Taiwo E. Adeleke1
This study examined the influence of artificial intelligence (AI) readiness on non-academic staff's skill rebalancing at the University of Uyo, Nigeria. Specifically, the study investigated the levels of skill rebalancing and AI readiness as well as the relationship between AI readiness (in terms of data, technological infrastructure, and AI policy) and non-academic staff's skill rebalancing at the University of Uyo. Two research questions and two hypotheses were developed to guide the study. A descriptive survey research design was adopted. The population comprised all senior non-academic staff of the University of Uyo. A sample of 731 respondents was purposively selected for the study. Data were collected using a researcher-developed instrument entitled Artificial Intelligence Readiness and Non-Academic Staff's Skill Re-Balancing Questionnaire (AIRNASSRQ). The instrument was validated by experts in Educational Management and Measurement and Evaluation, while its reliability was established using Cronbach's Alpha, yielding a coefficient of 0.71. The results revealed a moderate level of skill rebalancing among the non-academic staff at the University of Uyo (mean = 2.67, SD = 0.840), while the level of artificial intelligence readiness was low (mean = 2.38, SD = 0.65). Also, a significant positive relationship between artificial intelligence readiness and non-academic staff's skill rebalancing was found in the understudied university. The multiple regression analysis revealed that the dimensions of artificial intelligence readiness jointly predicted the non-academic staff's skill rebalancing significantly. The study concluded that artificial intelligence readiness is a significant determinant of non-academic staff's skill rebalancing in universities. It was therefore recommended, among others, that the university management should strengthen AI-related capacity-building programmes, improve digital infrastructure and internet connectivity, and enable non-academic staff to adapt effectively to the changing technological demands of university administration.
A study of Tunisian teachers' perceptions through a storytelling approach
Lilia Smida1; Ibticem Elhaj Fraj Ben Zammel2; Jamel-Eddine Gharbi3
The transition from traditional education to sustainability-oriented education promotes creativity, critical thinking, and human development. Integrating critical thinking across teachers, students, and content is essential. Hybrid models combining artificial intelligence tools with human teaching, as complementary pedagogical approaches, are recommended to enhance learning outcomes and critical thinking (Fakour & Eimani, 2025) while preserving human and emotional support (Smida, 2026). Strengthening digital literacy and critical thinking is therefore crucial for educators to adapt to AI-driven transformations in education. In examining the role of teachers in the era of AI integration, AI is perceived by most participants as a facilitator that enhances teachers' efficiency by providing educational resources and automating routine tasks, thereby reducing time-consuming activities and enabling educators to focus on higher-value pedagogical functions. However, unequal access to AI between teachers and students represents a significant risk that may reinforce existing educational inequalities. Consequently, universities have a responsibility to address these disparities and ensure that AI integration promotes equity rather than exclusion (Hmissi, 2026). Teachers' perceptions of AI in education vary according to their level of familiarity with the technology. Those with greater knowledge of AI tend to view it as a transformative educational tool, whereas less-informed teachers are more likely to express concerns and reservations (Laouel, 2025). Similarly, the use of AI among teachers at the Higher School of Design Sciences and Technologies in Tunisia is unequal, with only some integrating it effectively into their pedagogical practices. This disparity is mainly explained by differences in technological familiarity, limited training, and the absence of clear institutional frameworks. As a result, AI adoption ranges from basic supportive use to more advanced pedagogical integration (Fatnassi, 2026). These disparities exist not only within the same institution but also across different universities. Exploring differences between private and public higher education institutions may provide novel insights into AI adoption in higher education, particularly through a storytelling approach (Leah et al., 2010), revealing similarities and differences in teachers' familiarity with AI use and AI integration.
Constrained, collective, and improvised forms from the Global South
Jacob Filgona1; John Sakiyo2
Dominant theories of teacher agency have been developed primarily within Global North contexts characterized by relative resource adequacy, policy stability, and individual professional autonomy. Yet, the majority of the world's teachers, coming from the global south, work under conditions of chronic resource scarcity, policy incoherence, and systemic constraint—conditions these theories were never designed to explain. This paper addresses a central question: what forms does teacher agency take when these conditions do not hold? Drawing on literature from Nigeria, Ghana, South Africa, Zambia, and Kenya, this conceptual study develops a tripartite framework of agency in conditions of constraint. It conceptualizes improvised agency as real-time creative response when formal systems are inadequate, collective agency as forms of action distributed across professional networks rather than located in individuals alone, and constrained agency as professional action exercised within severe structural limitations rather than the absence of agency. The paper further proposes a dynamic relational model in which structural constraints generate conditions for improvisation, improvisational practices are sustained and refined through collective professional networks, and collective forms of action may, over time, contribute to the partial renegotiation of structural constraints. This framework builds on ecological accounts of teacher agency by shifting attention to how agency is negotiated in contexts marked by persistent scarcity and unstable institutional conditions. In doing so, it offers a lens for interpreting curriculum reform beyond stable policy environments. The paper concludes by considering the implications of this perspective for curriculum policy design, teacher professional development, and comparative studies of educational change.
A study of Jharkhand universities
Nilesh Kumar1
The use of Artificial Intelligence in education is changing the way we teach learn and run university system. Artificial Intelligence is making a difference in education all over the world. In India the government has introduced the National Education Policy 2020 which is pushing universities to use technologies and Artificial Intelligence in education. This study looks at how ready faculty members are to use AI tools in colleges and universities in Jharkhand. The research is trying to figure out how much faculty members know about AI tools, what skills they have and how they feel about using Artificial Intelligence in their teaching. It also tries to find out what problems faculty members are facing, like not getting training not having good enough computers and equipment not having clear instructions from universities. The research employs a quantitative methodology utilizing structured surveys through questionnaires distributed to faculty members from chosen universities and colleges in Jharkhand. Statistical methods will be employed to examine the relationship between faculty preparedness and variables like age, teaching experience, digital competency, institutional backing, and previous encounters with AI technologies. The study's findings aim to offer important insights for policymakers, university leaders, and educational planners in creating effective faculty development initiatives. The research will additionally add to the expanding academic literature on AI implementation in Indian higher education, especially concerning developing states such as Jharkhand.
Basil Osayin Daudu1; Salaudin Usman2; Michael Olanrewaju Oluyori2; Peter Anakobe Adinoyi3
In 2022, the National Universities Commission (NUC) introduced the Core Curriculum and Minimum Academic Standards (CCMAS), and its implementation in Nigerian universities started in 2023, replacing the outdated and rigid content-based curriculum, Benchmark Minimum Academic Standards (BMAS). However, educationists and non-educationists alike have concerns about the feasibility of CCMAS. This paper investigates whether CCMAS is outcome-based (problem-solving), as claimed by NUC, or content-based, like the old one. Therefore, using mixed methods, the study sampled university lecturers, students, and education analysts for this investigation because of their direct involvement and expertise. The responses generated from these participants were analyzed and interpreted using analytical and statistical techniques. Therefore, among others, the paper recommends consistent review and monitoring of CCMAS compliance to ensure quality and practical knowledge in the best interest of higher education in Nigeria.
A narrative review of teacher readiness
Dada Tolulope Racheal1
The rapid global adoption of artificial intelligence (AI) in education offers transformative opportunities to enhance pedagogical practices and optimise student learning outcomes. However, successful AI integration in secondary school classrooms fundamentally depends on teachers' readiness to adopt and apply these technologies effectively. This narrative review systematically synthesises the existing scholarly literature on teacher preparedness for AI integration in Nigerian secondary schools. Through a comprehensive analysis of 28 peer-reviewed empirical studies published between 2015 and 2025, the review establishes that Nigerian secondary school teachers demonstrate foundational awareness of AI concepts and tools, yet exhibit critical deficiencies in practical implementation skills. Key barriers include chronic shortages of reliable technological infrastructure, restricted access to structured professional development programmes, and the complete absence of national or institutional policy frameworks to guide AI adoption (Sambo-Magaji et al., 2025; Eke, 2024). The review further identifies urgent intervention priorities: mandatory AI-literacy training modules tailored to subject-specific pedagogy and the urgent formulation of clear, enforceable policy guidelines at both institutional and national levels. By clarifying the multifaceted dimensions of teacher readiness, this paper delivers evidence-based recommendations for policymakers, school administrators, and teacher educators to achieve equitable and sustainable AI integration in Nigerian secondary education.
An activity theory perspective
Beibei Ren1
Writing strategy research has undergone an epistemological turn from cognitive to sociocultural in recent years. Activity Theory (AT) represents a powerful framework to examine writing strategies from a sociocultural perspective. Though AT has been applied to investigate strategies writers employ in various written genres, how pre-service teachers employ strategies when writing Teaching Philosophy Statements (TPSs) for the first time has yet received scholars' attention. To address the research gap, this study conducts a qualitative study probing the technology-related mediating strategies in EFL pre-service teachers' TPS writing. The study collects multiple sources of data, including nine TPSs written by nine EFL pre-service teachers (one from each) and one-on-one interviews with the participants. As one variety of artefact-mediated strategies, a range of technology-related strategies are employed by the pre-service teachers. First, findings show that the pre-service teachers resort to the Internet, including such platforms as Google and YouTube, to gather information related to the TPS genre, which mediate their conceptualization of the TPS. Findings also demonstrate that the participants employ different kinds of technology-assisted writing tools when writing the TPSs, including using Youdao for vocabulary issues, Grammarly and Quillbot for grammar-related issues, and DeepL to help generate natural English expressions. This study yields important insights regarding EFL students' orchestration of technology-related writing strategies when they are novice writers of the TPS genre.
Opportunities, challenges, and institutional implications of AI tools in classroom teaching
Houda Fadel1
The rapid proliferation of Artificial Intelligence tools in educational settings is fundamentally reshaping the landscape of classroom teaching. From intelligent tutoring systems and automated feedback platforms to generative AI assistants, these technologies are no longer peripheral add-ons but increasingly central to how instruction is designed, delivered, and experienced. This paper offers a theoretical and conceptual examination of the integration of AI tools in classroom teaching, with particular attention to its pedagogical, ethical, and institutional dimensions. Drawing on contemporary educational theory — including constructivist and socio-cultural frameworks — this paper interrogates how AI tools alter the traditional roles of teacher and learner. It argues that while AI can personalize learning experiences, scaffold complex tasks, and reduce administrative burdens on educators, its uncritical adoption risks undermining core pedagogical values such as critical thinking, learner autonomy, and meaningful human interaction. The paper further explores the institutional challenges that arise as schools and universities attempt to incorporate AI into their curricula and governance structures. It examines tensions between innovation and regulation, equity of access, and the need for coherent institutional policies that balance technological affordances with educational integrity. Finally, the paper proposes a conceptual framework for what may be termed "AI-conscious pedagogy" — an approach that positions the educator not as a passive user of AI tools, but as a critical mediator who thoughtfully integrates technology in alignment with clear learning objectives and ethical responsibilities. This contribution aims to advance the theoretical discourse on AI in education and to inform policymakers, curriculum designers, and practitioners navigating this transformative moment in the history of teaching.
Mahona, Paschal Joseph1; Cynthia Adaeze Onuegwunwoke2
This study examined the transformation of teacher education in Africa within the context of the growing integration of artificial intelligence (AI) technologies into teaching and learning processes. Focusing on Nigeria and Tanzania as comparative case studies, the paper explored how AI-driven tools reshaped pedagogical practices, teacher preparation, professional identity, and ethical responsibility in higher education. The study was anchored on the Technological Pedagogical Content Knowledge and Sociocultural Theory, which provided theoretical explanations for technology integration, collaborative learning, and culturally responsive pedagogy in contemporary educational environments. The population of the study comprised student teachers and teacher educators from selected teacher education institutions in Nigeria and Tanzania, including Alvan Ikoku Federal University of Education, Owerri, Imo State, Nigeria. The paper adopted a qualitative-comparative approach to investigate the opportunities and challenges associated with AI integration in teacher education. The findings highlighted the pedagogical benefits of AI like personalized learning, instructional support, digital literacy development, and enhanced access to educational resources. Simultaneously, the study interrogated critical concerns relating to ethical literacy, technological inequality, overdependence on automated systems, and the preservation of human-centered teaching values within African educational contexts. The paper argued that effective teacher education in the AI era required a balanced framework that combined technological competence with ethical consciousness, intercultural awareness, and contextual responsiveness. The study concluded that African universities needed to strategically redesign teacher education curricula to prepare educators who could responsibly and critically engage with AI-driven educational transformation.
Challenges and issues for North African universities
Aziz Hantem1
Artificial intelligence (AI) has experienced remarkable growth driven by three converging factors: increasing processor power, the expansion of big data, and continuous algorithmic improvement (Portnoff & Soupizet, 2018; Ndione, 2022). Technologies such as machine learning, computer vision, and natural language processing (NLP) have progressively transformed many sectors. However, while the private sector has capitalized on these advances, public institutions particularly higher education establishments in developing countries, still lag significantly behind in AI adoption (Fokou, 2021). This paper examines the integration of AI in public management, with a focus on North African universities. It pursues two objectives: first, to identify the key challenges and barriers to AI adoption within these institutions; second, to analyze the causal relationships between AI adoption and improvements in university performance, both in terms of managerial governance and pedagogical innovation. The paper also proposes a conceptual model linking the determinants of AI adoption to organizational performance outcomes. The central research question is: to what extent does AI adoption contribute to improving the performance of North African universities, and what are the main facilitating and inhibiting factors? Theoretically, this research draws on the Technology Acceptance Model (TAM) and the Unified Theory of Acceptance and Use of Technology (UTAUT). Methodologically, a qualitative exploratory approach is adopted, combining a systematic literature review with participatory observation.
Transforming pedagogical beliefs and enhancing learner performance in the Algerian higher education context
Houyem Gasmia1; Farida Sadoune2
The rapid proliferation of artificial intelligence and digital technologies has fundamentally redefined expectations for language teachers worldwide. The capacity of English as a Foreign Language (EFL) teachers to integrate information and communication technologies (ICT) into their practice has emerged not merely as a professional enhancement, but as a prerequisite for pedagogical relevance in AI-mediated environments. This imperative is particularly pressing in Algeria, where systemic constraints have historically limited teachers' exposure to technology-enhanced professional development. This paper examines, from a theoretical standpoint, the conditions under which ICT-based in-service EFL teacher training may effectively transform pedagogical beliefs and promote learners' academic performance within Algerian higher education. Drawing on an integrated multi-theoretical framework — encompassing the Technological Pedagogical Content Knowledge model (TPACK; Mishra & Koehler, 2006), the Technology Acceptance Model (TAM; Davis, 1989), Vygotsky's sociocultural theory (1978), and Ertmer's first- and second-order barrier framework (1999, 2010) — the paper explores how these perspectives illuminate the mechanisms through which teachers construct and revise their beliefs about technology integration. It further argues that ICT-based training — when designed to address both infrastructural and belief-based barriers — holds significant potential for repositioning EFL teachers as agentic adopters of digital and AI-enhanced pedagogies. This argument is situated within broader discourses on digital equity, institutional change, and the ethical governance of AI in education, with particular attention to MENA and Global South contexts where such transformations remain undertheorized. Implications for EFL curriculum reform, teacher education policy, and the responsible integration of AI in language education are critically examined.
Constantine Passaris1
Artificial Intelligence (AI) represents a transformational innovation with profound implications for higher education. Since time immemorial higher education has played a foundational role in disseminating knowledge and empowering scientific advancement. AI has emerged as a transformative force in all fields of human endeavour and especially in higher education. Academic institutions worldwide must create new pathways and novel pedagogical paradigms by integrating AI into teaching, learning, scientific research, and academic governance models. In consequence, redefining the academic ecosystem for a new era with innovative technological enhancement. AI is the contemporary phase of internetization. Internetization is a new word and concept that I have coined to describe the electronic empowerment and global outreach of the 21st century. This paper will examine AI's impact on pedagogy, research, and academic governance. It will propose a framework for AI's responsible integration into academic institutions. In effect, mapping a road map that creates a blueprint for how AI can enhance teaching and learning, accelerate scholarly inquiry, and invigorate institutional governance. The paper's overarching conclusion is that higher education must embrace AI as a catalyst for innovation while proactively addressing risks related to academic integrity, algorithmic bias, and the need for ethical guardrails that safeguard human centered learning.
Roland Berberich
UK Higher Education Institutions (HEIs) find themselves in a seemingly existential crisis. Facing a spiralling decline in international recruitment already, the advance of generative AI (GenAI) disrupts and threatens traditional HEIs. Institutions and leaders struggle to a consistent and coherent response. Reactive "more of the same" or aggressive cost-cutting and staff reductions, are simply admitting defeat. Simultaneously deflecting accountability onto allegedly unforeseeable global instability and obstructive policymaking does not address the problem of a value proposition undermined by reality. This conceptual paper examines the underlying structural nature of this crisis and its impact on UK HEIs. Moving beyond superficial defence, this paper argues that while macroeconomics and geopolitics have undeniably disrupted domestic campuses, the true vulnerability lies with denial and resistance to change. Absence of strategic thinking on matters of internationalisation, institutions devoid of bold leadership became exposed to a perfect storm. Failing to proactively engage with GenAI and human intelligence, holding on to outdated concepts combined with a rapidly unravelling prerequisite for current internationalisation patterns HEIs find themselves in a desperate struggle for survival. The paper concludes offering a series of interventions for UK HEIs. Institutions must completely rethink their core value proposition. The future of higher education is argued to be surprisingly (non-)artificial. Survival hinges on a deliberate return to deep, human-centric, and relational education combined with sound strategy and decision making. Cultivating uniquely human capacities such as ethical judgment, criticality, reflexivity, and collaborative development universities can reclaim their distinct value in a world increasingly shaped by AI.
How AI personalization shapes university image and student choice
Shiva Rani1
For the past decades, universities have been increasingly inviting public figures, industry leaders, influencers, motivational speakers, celebrities, and entrepreneurs. The aim behind that university is to develop the brand, increase visibility, global ranking and student engagement. Specifically, before COVID, the university focused on academic quality, branding, internationalization and student experience. However, after the pandemic Events, institutional digital visibility, experiential marketing and reputation are increasing. The current study is based on the S-O-R model, which represent there is a stimulus that leads to the organism and response. In the current study, the AI personalize mega event act as a stimulus, student perception as organism response lead to university choices. The objective of this paper is to investigate the impact of AI personalization event and mega events on perception. Another objective is to investigate the impact of perception on university choices. The study was conducted in the Indian Context, specifically in the Uttarakhand context. The study used a quantitative research design and stratified purposive sampling. The study used 12th, graduate and post-graduation students of the last year. The study used a priori G Power software for sample size measurement. The study used partial least squares-structured equation modelling. The study found that there is a positive relationship between the variables. The study found that the perception has a partial mediation relationship between mega event and university choices. This study is play significant role in higher education marketing.
A performance analysis of top 25 NIRF 2025 universities
Debdas Mondal1
This study analyses the research performance, collaboration patterns, and academic impact of the Top 25 universities ranked in NIRF 2025 using multiple indicators derived from Scopus and Web of Science databases for the period 2020–2024. The analysis examines research output, citation impact, h-index values, international collaboration, funding patterns, subject distribution, and highly cited papers to understand the research landscape of leading Indian universities. The findings show that institutions such as the Indian Institute of Science, University of Delhi, Banaras Hindu University, and Jadavpur University dominate in terms of publication volume, citations, and overall research influence. Public universities constitute the majority of the top-ranked institutions, although private universities such as Manipal Academy of Higher Education, Birla Institute of Technology and Science Pilani, and Vellore Institute of Technology demonstrate increasing research productivity and international collaboration. The results further reveal that research output is largely concentrated in engineering, natural sciences, and medical sciences, while social sciences and humanities account for a smaller share of indexed publications. International collaboration, particularly with institutions in North America and Europe, contributes significantly to higher citation impact and the production of highly cited papers. Overall, the study highlights the importance of research productivity, collaborative networks, and global engagement in strengthening the academic performance and international visibility of India's leading universities.
A comparative study of institutional EMI policy and teacher agency in the MENA region and Global South
Soundous Benkorichi1
The adoption of large-scale educational policies often prioritizes measurable performance indicators over the complex, localized realities of classroom pedagogy. This study investigates the institutional implementation of English-medium instruction (EMI) in Algerian higher education, providing a critical MENA-region perspective on the global 'Internationalization of Higher Education' (IoHE) movement. Drawing on in-depth qualitative interviews with ten university instructors, this research identifies a persistent 'administrative-pedagogical disconnect.' Findings reveal that when institutions focus primarily on 'rushed' implementation and bureaucratic reporting, they trigger a 'tension gap' that neglects both student engagement and the professional expertise of faculty. This study explores the cognitive and professional negotiation processes instructors undergo when caught between globalized policy demands and the immediate, linguistic-pedagogical requirements of the classroom. Ultimately, this research argues that current institutional adoption models—often mirrored in the Global South—frequently overlook the essential 'human layer' of education. By highlighting the friction between top-down governance and bottom-up instructional enactment, the paper offers a critical critique of metric-driven reform. It concludes that sustainable educational policy must prioritize pedagogical sensitivity and teacher agency to ensure that institutional change translates into meaningful classroom success rather than mere administrative compliance.
Evaluating teaching quality and experience recognition in India's coaching-to-academia transition
Rohit Kumar Pal1
India's coaching industry has emerged as a significant parallel educational ecosystem, producing successful candidates in competitive examinations such as NEET, JEE, UPSC, and other national-level entrance tests. Many teaching professionals in this sector possess highly prestigious qualifications from institutions such as Indian Institutes of Technology (IITs) and Indian Institutes of Management (IIMs), alongside formal credentials including B.Ed and UGC NET/SET. Despite their proven teaching effectiveness and domain expertise, many are rejected from school, college, and university teaching positions due to limited recognition of coaching-sector experience. This study examines whether teaching quality can be measured solely through formal credentials such as B.Ed and NET/SET or whether practical teaching outcomes and experience should receive greater institutional recognition. The study analyzes data collected from 552 teachers and 337 educational administrators across India. Findings reveal that 45.1% of applicants were rejected despite substantial teaching experience, while administrative hiring policies remained heavily credential-oriented. Statistical analysis demonstrates a significant disconnect between practical teaching effectiveness and formal recruitment standards. Administrators acknowledged regulatory constraints linked to affiliation norms, API scores, and statutory compliance requirements. The research argues for a balanced evaluation framework integrating teaching outcomes, domain expertise, and professional experience alongside formal qualifications. The study contributes to policy discussions on teacher recruitment, educational quality assessment, and experience-equivalency frameworks within India's evolving education system.
Exploring techno-pedagogical competence, ethical awareness, and social responsibility in teacher education
Nazish Khan1
Artificial Intelligence (AI) is bringing major changes in the field of education. Teaching methods, assessment processes, classroom interaction, and digital learning practices are increasingly influenced by AI-based tools and technologies. In teacher education, these changes have created both opportunities and challenges for prospective teachers. The present study examines the role of professional development in improving techno-pedagogical competence and ethical responsibility among prospective teachers in the context of AI-supported education. The paper focuses on the need to prepare future teachers who can use technology effectively with proper pedagogical understanding and social responsibility. It explains that prospective teachers need continuous professional development to adjust to AI-based teaching tools, online platforms, and changing classroom practices. Along with technological skills, teachers also require ethical awareness, reflective thinking, decision-making ability, and responsible use of AI in educational settings. The study is conceptual and analytical in nature and is based on current discussions related to AI in teacher education and professional competency development. The paper states that teacher education programmes should give equal importance to technology, pedagogy, ethics, and human values. AI can support teaching and learning processes, but it cannot replace human interaction, emotional understanding, and professional judgement in education. The study adds to present academic discussions on AI, teacher education, and responsible educational practices, especially in the context of developing countries.
Boulmerka Soulef1
The current study investigates EFL teachers' attitudes toward Artificial Intelligence (AI) implementation at the Ecole Normale Supérieure (ENS) of Constantine, Algeria. The study highlights a tension between enthusiasm for perceived pedagogical benefits and apprehensions concerning ethical, technical, and institutional hurdles. This qualitative enquiry involved ten instructors chosen via purposive sampling, ensuring various expertise. Semi-structured interviews highlighted three principal themes: advantages (lesson design, engagement), ethical issues (privacy, over-reliance), and institutional limitations (lack of training/support). The findings highlight that for instructors to responsibly implement artificial intelligence (AI) tools in language classes, they require clear institutional regulations, ongoing professional training, and well-established ethical frameworks. While AI offers opportunities for personalized feedback and automated instruction, its integration introduces significant risks regarding data privacy, and the potential loss of human interaction in teaching.
Higher education students' views on AI, learning and the future of work
Pedro Pinheiro1
Artificial intelligence is transforming higher education and work, but students' perceptions of benefits, risks and responsible use remain uneven. This study examines how higher education students understand, use and evaluate AI tools in academic and professional contexts. Drawing on survey data from 104 respondents, mostly undergraduate students in management, business, economics and finance, the research explores AI experience, perceived impacts, academic uses, ethical concerns and readiness for AI-supported decision-making. The findings indicate a pragmatic but cautious attitude. Most respondents see AI as useful yet risky, and identify technology, marketing, education and finance as areas of major impact. While self-assessed knowledge of AI is moderate, students associate AI with workplace efficiency and support for learning, especially through research assistance, summarisation, idea generation, personalised learning and clarification of complex topics. At the same time, trust in AI for business decision-making and confidence in distinguishing generative AI from other AI systems are weaker. Respondents express concern about over-reliance, job loss, privacy, transparency and algorithmic bias. In educational terms, students support the integration of AI into university courses and strongly favour institutional guidelines for ethical academic use. The study contributes to interdisciplinary debates on AI in education, culture and society by showing that students are not simply enthusiastic adopters or resistant users. Rather, they combine frequent use with a demand for responsible governance, pedagogical guidance and critical AI literacy. These results suggest that higher education institutions should move beyond access to tools and develop curricula integrating technical, ethical and professional dimensions of AI competence.
Roland Berberich1
After panic-induced return to in person exams and increased AI policing, it appears Higher Education Institutions (HEIs)are ready to raise the white flag of surrender. No credible attempt appears to be made to genuinely question the value proposition of a university education in the 21st Century. While the silver lining of a Covid vaccine spurred vocal "back to the classroom" movements, the rise of Generative AI (GenAI) again sees HEIs ignoring their own teachings and increasingly desperate efforts to maintain the classroom and essay comfort zone for academics that have seemingly forgotten the crucial nature of their role. A global pandemic or technological advancements are not just disruptors or an inconvenience, they have become catalysts for cataclysmic change that forces HEIs at all levels to reframe their value to an increasingly critical audience. By illustrating the many failures of the higher education business this paper demonstrates that institutional disengagement, credentialism, and the commodification of degrees alienated the audience long before GenAI. The paper proposes to abandon assembly line transactionality where students rightfully utilise GenAI to fulfil bureaucratic mandates. Instead, a return to transformative experiences that embed GenAI as a component the learner can critically engage with. Continued deflection away from failed strategic leadership only delays rebuilding a transformative, collaborative scholarship experience which GenAI cannot simulate.
Samson Mutsagondo1
Zimbabwe unveiled its National Artificial Intelligence Strategy on 13 March 2026. The development was hailed as a significant step towards artificial intelligence (AI) adoption in the country. This study interrogates the impact of the national strategy on AI adoption in records management in Zimbabwe. The study examines the contribution of each of the six core pillars of the strategy in relation to AI adoption in records management. The pillars are talent development, infrastructure, service transformation, governance, research, and strategic collaboration. The strategy lays a solid foundation for AI adoption and implementation. In addition, it enhances technological sovereignty and the protection of domestic data assets, and seeks to develop national infrastructure, skills and collaborations, from which records and archival institutions may greatly benefit. These factors are set to rescue Zimbabwe from the low AI adoption and implementation tier on the world AI readiness and adoption rankings. Nonetheless, the study reveals many implementation gaps owing to the discrepancy between theory and practice. Among the national strategy's shortfalls in records management are skills deficiencies, regulatory lapses, low capital investment, chronic power shortages, and tight fiscal constraints, thus making a mismatch between strategic ambition and execution capacity. The study serves as a critical reminder of the importance of political will and commitment, and resource mobilisation and alignment for the country to make meaningful strides in AI adoption and implementation. Data in this study were obtained from the review of literature and documents on AI, national AI strategy and policy and AI adoption and implementation.
Incorporating AI into journalism school curricula
Shawn McIntosh1
Digital media has brought disruptions for decades now to the journalism profession and the news industry, and AI promises—or threatens—even more disruptions. Nevertheless, AI tools are being increasingly incorporated into newsroom workflows even as the ethics of AI and best practices using AI are very much still evolving. The challenges of AI are compounded in an educational setting, where students may frequently use AI in courses, and may not have the same negative perceptions around its use that faculty (and some professionals) may see. This paper addresses the difficult and delicate balancing act that journalism schools face in teaching and preparing journalism students to utilize AI to do better journalism while not undermining the fundamental critical writing, thinking, and analytical skills journalists increasingly need to do their jobs well. The paper argues that it is an educational disservice for journalism schools to ignore or downplay the role of AI in the profession, as students must be prepared for a world in which AI tools are widely used. Furthermore, without giving students the faculties by which to critically examine how AI is used, the effects of AI on their work, how AI affects public trust in journalism, and how AI will further change the professional roles of journalists themselves, it is likewise leaving them unprepared when entering the profession. The paper explores ways this educational balance may be achieved by examining some successful cases of journalism schools incorporating AI into their curricula.
Artificial intelligence as a catalyst for CLIL transformation in Algerian higher education
Benabdelouahab Amina1; Amrate Mustafa2
Content and Language Integrated Learning (CLIL) has gained attention in higher education due to its capacity to develop disciplinary knowledge and English language competence within an integrated pedagogical framework. In the Algerian context, implementation faces structural and pedagogical constraints that shape classroom realities. These constraints include uneven linguistic proficiency among learners and limited instructional resources, alongside pressures linked to academic internationalization. Recent advances in Artificial Intelligence (AI) introduce for pedagogical innovation and curriculum enhancement. Moreover this study examines potential of AI as a catalyst for transforming CLIL practices in Algerian higher education. A qualitative exploratory design guides process to capture academic experiences. Data were collected through semi-structured interviews with fifteen university teachers engaged in English-medium instruction and CLIL-related practices at an Algerian university. Thematic analysis structured data and highlighted perspectives, opportunities, and challenges related to AI integration in CLIL contexts. Furthermore findings indicate that AI-based tools extend pedagogical scope of CLIL in ways. They support adaptive learning pathways and provide immediate feedback systems. They also offer instructional support and refined assessment practices. Participants describe how these tools strengthen learner autonomy, enrich classroom interaction, and facilitate engagement with disciplinary content. Additionally the analysis further shows that AI contributes to reshaping institutional practices and strengthening academic quality while supporting educational reform agendas. Ethical concerns, digital infrastructure limitations, and teacher preparedness emerge as conditions shaping implementation. The study positions Artificial Intelligence as a catalyst for the evolution of CLIL ecosystems in Algerian higher education and highlights implications for future pedagogical development.
A workshop model for responsible self-presentation
Kai Löser1
Application processes are increasingly shaped by AI on both sides: organizations use automated tools to screen, rank, or evaluate applicants, while applicants themselves use chatbots to draft their materials. This dual AI mediation creates a pedagogical problem for career guidance: students need to understand how they may be assessed while learning to use AI without surrendering privacy, integrity, or their own voice. This contribution examines "Bewerben mit KI" (Applying with AI), a three-hour workshop developed at the University of Rostock and proposed for adaptation to study and career guidance in schools. The workshop offers a practice-based design for AI literacy in application coaching. It introduces generative AI as probabilistic text production and treats applications as situated self-presentation rather than template completion. Participants analyze job advertisements, translate competencies into biographical evidence, compare generic and personalized cover-letter drafts, and simulate interview situations. Each step includes critical checks: What became clearer? What became exaggerated or generic? Which data should not be shared? When does assistance become misrepresentation? In a final reflection, participants formulate a personal rule for AI use: what they delegate, what remains their own, and how privacy and integrity are protected. Transferring the format to secondary schools would require age-appropriate moderation, data minimization, and institutional safeguards. The contribution argues that responsible AI use in career guidance is not a matter of better prompting alone, but of structured comparison, deliberate limits, and reflection.
Empirical evidence, competencies, training needs, and pedagogical challenges
Sophia Crescentini1; Marta Salvucci2
The rapid and massive integration of Generative Artificial Intelligence (AIG), particularly via user-friendly chatbot applications, are shifting human experiences into an "onlife" hybrid environment. In higher education, this widespread yet spontaneous adoption introduces important pedagogical challenges, including the risk of cognitive outsourcing, epistemic deprivation, and an inadequate conception of knowledge and learning. To explore the current state of AI literacy, the contribution presents an empirical study conducted at the University of Urbino Carlo Bo. The study utilized an adapted Italian version of the standard SDSU (San Diego State University) AI Survey, specifically analyzing responses from 38 professors and 74 students regarding competencies and institutional training. The findings reveal a significant cultural and institutional gap. While 65% of students feel confident using AI, their security is tied to practical usage rather than algorithmic understanding. Conversely, only 30% of professors feel capable of explaining basic AI mechanics, and 72% report a profound lack of adequate institutional training. Furthermore, 82% of students declare to operate in a normative "grey area" without clear guidelines. Despite these shortcomings, strong interest exists for future integration: 88% of students demand practical AI workshops for their professional curricula, and 60% of faculty support mandatory training modules, while expressing concerns over increased workloads. Bridging this gap requires moving beyond mere instrumental mastery toward a comprehensive "AI literacy" framework. In particular, the paper advocates for robust Faculty Development programs and innovative instructional designs centered on human-machine collaboration and aimed to foster higher-order cognitive skills, ethical-intellectual virtues, and a flexible mindset.
Saadaoui Marouane1
The rapid integration of artificial intelligence (AI) into educational systems worldwide has opened new avenues for transforming teaching practices, learning environments, and institutional performance. This study investigates the impact impact of AI adoption and pedagogical innovation on the efficiency and quality of the Moroccan educational system . Drawing on an interdisciplinary framework combining educational sciences, economics of education and digital sociology , this research addresses three core questions : How are AI- based tools currently being adopted in Moroccan educational institutions ? To what extent does pedagogical innovation mediate the relationship between AI integration and educational outcomes ? And what structural,cultural, and institutional barriers constrain this transformation in a developing-country context ? Using a mixed-methods approach including a systematic literature review, comparative analysis of MENA educatonal policies, and a field survey targeting students and faculty at Sultan Moulay Slimane University , the study identifies critical succes factors and persistent obstacles to effective AI integration , Findings highlight significant disparities in digital infrastructure, uneven teacher digital literacy, and the absence of a coherent national AI in education strategy as key limiting factors . This paper contibutes to the growing scholarship on AI driven educational reform in the Global South and offers evidence-based policy recommendations for Moroccan decision-makers, educators and academic institutions commited to buildin a more equitable, innovative and performance-oriented educational system .
An investigation of EFL teachers' attitudes
Aoula Cherbal1
The artificial intelligence (AI) rapid proliferation has affected the realm of education ushering in radical transformations and reshaping instructional demands. In a multilingual setting, AI literacy has emerged as an essential competency. It enables teachers to effectively utilize technological resources and enhance their professional skills as it supports innovative teaching practices, personalized learning, and improved classroom engagement. Yet, many teachers often fail to interact meaningfully outside the boundaries of their conventional methods and encounter difficulties in applying AI tools. This empirical study aims to investigate university teachers' attitudes towards the effectiveness of AI literacy on the enhancement of their teaching skills and professional growth. A mixed-methods employed a sequential exploratory design was adopted. Quantitative data were collected through a questionnaire administered to 50 EFL teachers from the Department of English at the University of Jijel. To gain deeper insights into participants' perceptions and experiences, qualitative data were gathered through semi-structured interviews. The quantitative data were analyzed using descriptive statistics, while the qualitative data were examined through thematic analysis. The findings reveal that teachers generally held positive attitudes towards AI literacy and recognize its potential to improve instructional planning, classroom management, assessment practices, and professional decision-making. Classrooms led by AI trained teachers were perceived as inclusive, culturally sensitive, and reveal students' engagement. This study addressed the human element by showing how AI literacy is perceived as central for students' engagement and teachers' agency. It contributes to pedagogy by providing insights about the challenges encountered and the importance of innovative pedagogical practices for classroom interaction and teacher professional development programs. It informs policy- makers how teachers' training in AI can bring technical skills and empowerment when appropriately applied in multicultural contexts.
A perspective
Nagendra Kumar1; Mahesh Krishna2
Artificial intelligence in the Global South raises a question that is also a provocation: who builds knowledge, and under what material conditions? India sits somewhat awkwardly in this picture. Its institutional ambitions are real (the India AI Mission's cabinet-approved ₹10,372 crore outlay, AICTE's declaration of 2025 as the Year of Artificial Intelligence across 14,000+ institutions), but ambition and structural position are different things, and India's structural position remains that of a country whose AI ecosystem depends heavily on tools built elsewhere, for other people. This paper examines AI adoption in Indian higher education through that lens. Drawing on NEP 2020, the India AI Mission, AICTE mandates, and recent survey data on institutional uptake, it argues that India's current trajectory reproduces existing hierarchies of knowledge production. The dependence on anglophone platforms (OpenAI, Google, Microsoft) trained on data that skews toward Western epistemological norms is the clearest instance of this. An AI infrastructure that cannot adequately process Hindi, Tamil, or Bengali is doing something more consequential than failing a usability test; it is, in Santos's terms, a vector for epistemicide. The paper also takes seriously the material conditions that underpin this. The digital divide between urban and Tier 2/3 institutions, the documented unpreparedness of faculty (only 17% qualify as advanced AI users), and the data poverty of Indian HEIs are not incidental frictions in an otherwise sound policy framework. They reflect a deeper problem: that India's AI ambitions are being pursued within an architecture never designed with the Global South in mind. The paper ends with a proposal for what equity-first, epistemically sovereign AI policy in Indian higher education would actually require, with Indic-language model development and community-grounded data governance at its centre.
Ethical judgment and professional agency in higher education
Sondes Tamer1
The rapid integration of Generative Artificial Intelligence (GenAI) into higher education has fundamentally disrupted traditional pedagogical models, necessitating a profound re-evaluation of teaching competence. This article explores how the shift from knowledge transmission to AI- mediated learning environments redefines the professional role of the educator. It argues that technical proficiency in AI is insufficient; rather, the core of modern teaching competence lies in the exercise of ethical judgment and the assertion of professional agency. It aims to highlight the transition of teachers/educators from content providers and instructors to "learning architects and critical evaluators" of AI generated content. Adopting a descriptive approach, data were collected through interviews, and questionnaires distributed to 15 EFL teachers of Chadli Benjedid University (CBU) El Tarf to gain deeper understanding to their experiences. Findings suggest that the future of higher education depends on the ability of teachers to reclaim and empower their agency as critical curators of technology, ensuring that AI enhances rather than replaces the essential human dimension of the teaching-learning relationship.
A systematic literature review
Malak Gattaf1
The rapid advancement of artificial intelligence (AI), particularly generative AI and large language models, has transformed higher education and heightened the need for English as a Foreign Language (EFL) teachers to develop AI literacy. AI literacy encompasses the knowledge, skills, attitudes, and ethical awareness required to critically understand, evaluate, and responsibly integrate AI technologies into teaching and learning practices (Long & Magerko, 2020; Ng et al., 2021). In higher EFL education, AI-powered tools are increasingly employed for lesson planning, content generation, assessment, feedback, and personalized learning, underscoring the importance of preparing Algerian EFL teachers to effectively navigate AI-mediated learning environments. This paper presents an ongoing systematic review of teacher AI literacy in higher EFL education, with particular attention to its implications for Algerian higher education. Guided by the PRISMA framework, relevant studies are being identified through major academic databases and analyzed thematically. The review focuses on conceptualizations of AI literacy, key competencies required for EFL teachers, teachers' perceptions and readiness, professional development initiatives, and challenges associated with AI integration. Preliminary findings suggest that EFL teachers generally recognize the pedagogical potential of AI technologies and their capacity to enhance teaching practices and learner engagement (Kasneci et al., 2023). However, emerging evidence points to persistent challenges related to limited AI competencies, insufficient training opportunities, ethical concerns, and the lack of discipline-specific AI literacy frameworks. By mapping current research trends and identifying gaps in the literature, this review seeks to provide insights for teacher education, institutional policies, and professional development programs aimed at fostering the responsible and effective integration of AI in Algerian higher education.
A systematic review of teacher AI literacy in higher education
Malak Gattaf1
The rapid advancement of artificial intelligence (AI), particularly generative AI and large language models, has transformed higher education and heightened the need for English as a Foreign Language (EFL) teachers to develop AI literacy. AI literacy encompasses the knowledge, skills, attitudes, and ethical awareness required to critically understand, evaluate, and responsibly integrate AI technologies into teaching and learning practices (Long & Magerko, 2020; Ng et al., 2021). In higher EFL education, AI-powered tools are increasingly employed for lesson planning, content generation, assessment, feedback, and personalized learning, underscoring the importance of preparing Algerian EFL teachers to effectively navigate AI-mediated learning environments. This paper presents an ongoing systematic review of teacher AI literacy in higher EFL education, with particular attention to its implications for Algerian higher education. Guided by the PRISMA framework, relevant studies are being identified through major academic databases and analyzed thematically. The review focuses on conceptualizations of AI literacy, key competencies required for EFL teachers, teachers' perceptions and readiness, professional development initiatives, and challenges associated with AI integration. Preliminary findings suggest that EFL teachers generally recognize the pedagogical potential of AI technologies and their capacity to enhance teaching practices and learner engagement (Kasneci et al., 2023). However, emerging evidence points to persistent challenges related to limited AI competencies, insufficient training opportunities, ethical concerns, and the lack of discipline-specific AI literacy frameworks. By mapping current research trends and identifying gaps in the literature, this review seeks to provide insights for teacher education, institutional policies, and professional development programs aimed at fostering the responsible and effective integration of AI in Algerian higher education.
Reem Khalouk Issa1
This study is concerned with ELT in higher education. It aims at examining the issue of how we can achieve more effective teaching/learning in the field of English language teaching through exploring teachers' and students' perceptions. The study presents a comparative analysis of students' and teachers' perceptions regarding effective teaching in the context of Aleppo University- specifically at the Higher Institute of Languages. It employs semi-structured interviews with six teachers and six students. The findings show that students perceive effective teaching as teaching in a clear and easy language; they highlight similarities and differences in students' and teachers' perceptions. The findings also refer to the factors that influence such perceptions such as educational and personal factors. The findings suggest and recommend diverse teaching strategies and techniques to be used in the field of ELT. They also yield practical implications for teachers and teacher trainers.
N. Raghavendra Rao1
The basic functions, such as production, marketing, and finance, have remained the same as the world moved from an agricultural economy to an industrial economy. Business processes, procedures, methods, strategy, management thinking, and approaches related to the basic functions have been changing. Any change, such as adding new features to existing products, designing a new product, changing business processes, the business model, and other activities within the basic functions, has to be carried out. This assignment is given to the research and development department in an enterprise. The role of research and development in an enterprise is to think of a new approach for the above activities. If there is no research and development within an enterprise, a group of executives, at the instruction of top management, will form a team to implement an innovative approach for the above activities. Their innovative approach will be based on their professional experience and knowledge. Generally, their approach will aim to be competitive in the market. In the present business scenario, it is not enough for an enterprise to be competitive. It is required to be acceptable as a supplier or service provider. There is a need for a knowledge-based system to be created in an enterprise. Many intelligent software tools are required to make use of this system. These tools have only the specific features needed to use the data in this system. The concept of artificial intelligence enables the creation of the specific required features for an enterprise.
Opportunities, ethical challenges, and the future of higher education
Sukhmeet Kaur1
The rapid adoption of Generative Artificial Intelligence (GenAI) tools, such as ChatGPT, Claudie, and other LLM Models, is shaping higher education worldwide, including in India. At the same time, the rigorous use of AI raises concerns for the ethical use of AI and its future challenges. This conceptual paper reviews recent developments in the literature on Gen AI in Indian higher education and explores their implications in the context of the National Education Policy (NEP) 2020. The paper discusses the opportunities AI offers for personalized learning, improved academic support, and enhanced student engagement, while also highlighting challenges related to ethical use, overdependence on AI-generated content, and the digital divide. The paper concludes that a balanced and human-centered approach can help Indian universities harness the benefits of AI while preserving academic integrity and educational quality. The study contributes to the ongoing discussion on AI-enabled education and offers recommendations for policymakers, educators, and students in the Indian higher education context.
Artificial intelligence, information integrity and social responsibility in higher education
Edmont Pasipamire1
Artificial Intelligence (AI) is changing the way knowledge is created, accessed and evaluated in higher education. While artificial intelligence provides opportunities for learning and research, it also poses challenges in terms of misinformation, fake content, algorithmic bias, intellectual property and the responsible use of information generated by AI. This paper presents a scope review of research in the fields of artificial intelligence, academic libraries, information integrity and social responsibility in higher education. The review systematically maps existing literature to identify the main themes, emerging concerns and gaps in the knowledge on the role of libraries in the information environment mediated by artificial intelligence. Special attention is paid to AI literacy, critical information assessment, information integrity, ethical use of artificial intelligence and the evolving responsibilities of librarians. The paper conceptualises libraries as compasses in an information landscape mediated by artificial intelligence, helping learners and researchers navigate an increasingly complex information environment while maintaining critical judgment and ethical awareness. The review shows how the role of academic libraries extends beyond traditional access to information to promote responsible use of knowledge produced by artificial intelligence. Linking libraries and information science to AI-related discussions in education and society, the paper highlights the importance of information integrity and human-centred training in developing responsible AI-enabled learning environments.
Artificial Intelligence, Governance and the Future of Quality Assurance in the Global South
Dervin K. Cochrane1
Can artificial intelligence transform accreditation in the Global South from a periodic, compliance-driven exercise into a more transparent, continuous and integrity-focused system of higher-education governance? This paper addresses this question by examining the emerging role of artificial intelligence (AI) and digital technologies in accreditation against the structural vulnerabilities that characterize quality-assurance systems in Jamaica, the Caribbean and the wider Global South. Drawing on my recent broader interdisciplinary book chapter on accreditation, governance and AI, the paper argues that accreditation should not be understood merely as a technical mechanism for verifying institutional quality. It also operates as a system of governance, legitimacy and power. Yet its dependence on periodic reviews, institutional self-reporting, documentary evidence and centralized decision-making can create opportunities for symbolic compliance, data manipulation, regulatory capture and other forms of institutional gaming. These vulnerabilities are conceptualized through the “Accreditation Wall”: the assumption that formal accreditation procedures constitute an effective barrier against institutional failure and misconduct, even when actors may circumvent the safeguards themselves. The paper argues that Artificial Intelligence (AI) creates an opportunity to reconceptualize this model. AI-assisted anomaly detection, continuous data analysis, digital credential verification and real-time monitoring could enable accreditation systems to move beyond episodic procedural compliance towards more dynamic forms of institutional accountability. For resource-constrained systems in the Global South, these technologies could potentially strengthen regulatory capacity while improving transparency and the early detection of irregularities. However, technological adoption does not automatically produce better governance. Artificial Intelligence (AI) may reproduce existing inequalities and introduce new problems involving algorithmic bias, data governance, technological dependency, institutional capacity and accountability for automated decisions. The paper therefore rejects both technological determinism and the assumption that Artificial Intelligence (AI) can substitute for professional judgement. It concludes that the central question for the Global South is not simply whether AI should be incorporated into accreditation, but how AI can be governed so that it strengthens rather than reproduces the weaknesses of existing accreditation systems? A future accreditation architecture should consequently combine technological innovation with human oversight, transparency, contextual relevance, institutional integrity and social justice.
A Responsible Management Perspective
Hubert Gąsiński
The growing integration of artificial intelligence into the banking sector is transforming work organisation, managerial decision-making and employee responsibilities. Although AI can improve operational efficiency, support decision-making and facilitate knowledge management, its implementation may also generate new occupational challenges, including technological stress, uncertainty about professional roles, reduced autonomy and concerns about job security.
This paper aims to examine the relationship between AI-driven digital transformation, employee well-being and occupational stress in the banking sector, with particular attention to the role of responsible management practices. Building on research on employee well-being, organisational climate, organisational agility and the use of information and communication technologies, the study will explore how managerial support, employee participation, digital competencies and transparent communication may influence employees’ responses to AI adoption.
The paper will develop an analytical framework linking AI implementation, organisational practices and employee outcomes. An integrative review of the relevant literature will provide the basis for identifying key relationships and formulating propositions for future empirical research.
The study seeks to contribute to the literature on responsible digital transformation by highlighting the importance of balancing technological innovation with employee well-being, professional autonomy and sustainable organisational development.
Georgios Alexandropoulos
This project presents an AI-assisted educational digital game designed to support the teaching and learning of the Neolithic Age through an interactive and experiential environment. The game was developed using AI tools, including ChatGPT, to assist in content creation, question design, narrative structuring, and educational scaffolding aligned with pedagogical objectives. The virtual environment places students in a Neolithic village where they explore different areas such as agricultural fields, animal domestication zones, tool-making spaces, and communal areas. Through interaction with these spaces, learners answer multiple-choice questions related to key Neolithic developments, including early agriculture, domestication of animals, technological innovation, and social organization. The design is grounded in constructivist learning theory (Piaget), multimodal learning approaches (Cope & Kalantzis), and supported by elements of behaviorism, cognitivism, and social learning theory (Vygotsky). Gamification strategies such as immediate feedback, scoring systems, and progressive cognitive challenge are integrated to enhance motivation, engagement, and knowledge retention. The instructional structure also draws on established educational frameworks, including Bloom’s Taxonomy, Gagné’s Nine Events of Instruction, and the ADDIE model, ensuring systematic alignment between learning objectives and game mechanics. The game encourages active learning by requiring students to explore, interpret, and apply knowledge rather than rely on memorization. By combining historical content with interactive gameplay and AI-supported design, the project promotes critical thinking, digital literacy, and deeper conceptual understanding of Neolithic history. Overall, the educational game demonstrates how artificial intelligence can enhance instructional design and contribute to the development of engaging, student-centered learning environments.
A predictive model of analysis
Jyoti Kumari1; Atasi Mohanty1
Generative artificial intelligence (AI) tools are increasingly embedded in higher education, yet little is known about the psychological mechanisms linking AI-enabled learning experiences to student engagement and well-being. This study examined generative AI use as a predictor of academic engagement and well-being, testing emotional intelligence and motivation and resilience as parallel mediating mechanisms, and estimating a preliminary construct-level structural model of the broader hypothesized architecture. A quantitative, cross-sectional design was used with a convenience sample of 129 undergraduate and postgraduate students (aged 18-28) at the Indian Institute of Technology Kharagpur. Participants completed an online questionnaire assessing 13 constructs spanning AI-enabled learning support, academic engagement, psychological resources, and well-being. Regression-based mediation analyses, following Hayes' (2022) PROCESS macro logic with 5,000-sample bootstrapped confidence intervals, showed that generative AI use significantly predicted academic engagement and social and digital well-being. Emotional intelligence consistently mediated all three relationships, while motivation and resilience additionally mediated the pathways to social and digital well-being; mediation was partial for academic engagement and approached full mediation for the well-being outcomes. A complementary construct-level structural path model, estimated because item-level data were unavailable for most constructs, corroborated this pattern: generative AI use predicted instructional design quality, which in turn predicted academic engagement, emotional intelligence, and motivation and resilience, which subsequently predicted the well-being and lifelong-learning outcomes (β = .16-.56, all p < .05). Overall model fit fell short of conventional benchmarks (CFI = .81, TLI = .63, RMSEA = .26), indicating these structural estimates are provisional pending a full latent-variable confirmatory factor analysis. Psychological well-being was excluded throughout due to inadequate measurement of reliability. Findings suggest generative AI tools may support student engagement and well-being primarily by strengthening emotional intelligence and motivational resilience, offering a preliminary target for AI-integrated learning interventions in higher education.
Student trust, assessment integrity, and institutional governance in higher education
Arpan Shrivastava1; Gunjan Anand2; Arpita Shrivastava3
Artificial Intelligence (AI) is rapidly transforming higher education through adaptive learning systems, automated assessment tools, predictive analytics, and AI-supported pedagogical practices. While these technologies offer opportunities for personalized learning, instructional efficiency, and institutional innovation, their adoption also raises significant concerns related to assessment integrity, student trust, ethical governance, digital inequality, and institutional readiness, particularly within the Global South. Existing research on AI in higher education has largely focused on technologically advanced contexts, with limited attention given to the socio-economic, infrastructural, and governance challenges experienced by institutions in developing educational ecosystems. Addressing this gap, the present study examines the role of responsible AI practices in shaping educational transformation within higher education institutions in the Global South. Drawing upon the Technology Acceptance Model (TAM), Institutional Theory, and Responsible AI governance frameworks, the study develops a conceptual model investigating the relationships among AI transparency, institutional readiness, perceived fairness, AI literacy, assessment integrity, and student trust in AI-mediated educational environments. The study further explores how institutional governance mechanisms and ethical AI practices influence student engagement, perceptions of fairness, and acceptance of AI-supported assessment systems. A quantitative research design employing Structural Equation Modeling (SEM) is proposed to examine the hypothesized relationships using data collected from university students and faculty members across selected Global South higher education institutions. The study contributes theoretically by integrating responsible AI governance with educational transformation and student trust literature within underrepresented Global South contexts. Practically, the findings are expected to provide strategic insights for universities, policymakers, and educational administrators regarding ethical AI adoption, transparent assessment practices, faculty AI literacy, and inclusive institutional governance. The paper ultimately proposes a human-centered AI governance framework capable of balancing technological innovation with educational equity, trust, academic integrity, and sustainable institutional transformation in higher education.
Parvathi Devi Y1; Nayanasri Dusanapudi1; Vinothini P.S1; Geetha Mary Amalanathan1
The students of today encounter numerous sources of challenge, including academic demands, emotional fitness, and preparation for their profession. The result is that intelligent support systems need to be provided across a range of integrated interfaces. While there are numerous solutions that address portions of these challenges, they have typically been found to be supported via multiple signatures, and are often located between different platforms that make it difficult for the student to use each service in an efficient manner. This paper will present the Smart Student Suite as an Integrated and AI-enabled student support platform which addresses the four identified areas of support (Academic, Emotion, Career and Writing) as identified. The Academic Module will utilize predictive analytics to identify at risk students and to understand how they are likely to perform in the future. The Emotion Module will utilize multimodal analysis to help determine a student's emotional state and to provide appropriate support in a timely manner. The Writing Module will utilize advanced natural language processing techniques to support the student via support in the improvement of their writing structure, clarity and overall writing performance. The Career Module will also utilize trade fraud detection methodologies to identify unsafe job postings and also to ensure the student's access to legitimate opportunities. By co-locating these features into a single platform, the system is intended to provide a proactive and holistic support environment that will enhance student's academic performance, enhance student's mental well-being, and improve graduate career readiness.
A case study on opportunities and teachers' perceptions
Kenza Takarroucht
The rapid and steady technological advances have led to the incorporation of AI-powered assistants in nearly every pedagogical practice. This trend inevitably impacts assessment practices. However, many educators question the processes and the opportunities that are not adequately researched in the field of EFL teaching. Therefore, this study aims to explore how EFL teachers use AI for assessment purposes and their perceptions towards its use. To that end, the study employed an explanatory sequential mixed methods design. In the first phase, quantitative data were collected through a questionnaire and analyzed using correlational statistics to examine the association between AI use and assessment practices of a group of teachers (n=30). In the second phase, qualitative data were gathered through semi-structured interviews with selected participants to gain a deeper understanding of the quantitative findings and explore teachers' assessment practices. Data indicated that AI-assisted assessment was supportive in terms of generation of criteria, scoring, and feedback on the writing process. Moreover, teachers voiced positive perceptions with regard to reliability, while they raised concerns about validity.
Legal and ethical challenges in higher education
Megha Ojha1; Moutushi Thakur2
The fast integration of Artificial Intelligence (AI) into higher education has changed teaching, learning, assessment, and research practices. Innovative generative AI tools, such as large language models, automatic content generators, and intelligent tutoring systems, offer unprecedented opportunities to improve educational access, deliver more tailored learning, and boost research output. However, these technological advancements also raised major concerns about academic integrity, authorship, originality, responsibility, and ethical behaviour. The use of AI in academic and intellectual contributions challenges the traditional idea of plagiarism, research integrity, and authenticity. This paper examines the legal and ethical side of AI-assisted academic work in higher education and critically examines emerging challenges associated with AI-generated academic contributions. Through a doctrinal and comparative analysis, the paper highlights the need for balanced governance mechanisms that promote innovation while safeguarding academic morality and ethical standards. The paper argues that effective use of AI in higher education requires robust institutional policies, ethical guidelines, and effective assessment policies. The paper concludes with recommendations for educators, policymakers, and higher education institutions to foster responsible AI use while keeping the core values of academic integrity, fairness, and trust.
Meciad Meriem1; Boudechiche Hamid2
The swift expansion of generative artificial intelligence tools in higher education has fundamentally reshaped the landscape of postgraduate research. This evolution raises urgent questions about whether students have cultivated the ethical literacy necessary to responsibly adapt to these developments. As universities increasingly create guidelines about when it is acceptable to use AI in academic work, recent studies indicate that simply knowing these policies does not guarantee that people will act ethically. Another core issue is that many master students still lack a coherent understanding of what responsible AI use truly means within the specific demands of their dissertation research. This research explores the nature and depth of AI literacy among second-year master's students actively engaged in dissertation writing, focusing especially on the discrepancy between how familiar students claim to be with AI tools and their actual depth of understanding regarding the ethical implications. Using a questionnaire given to both second-year master students and their academic supervisors, the research highlights how each group receives ethical awareness, usage behaviors, and the adequacy of institutional guidance. By comparing students' and supervisors' views, it pinpoints where understanding fails and where governance measures are most urgently needed. The study contributes to an emerging body of scholarship on AI ethics in higher education by introducing an operational framework for AI literacy at the postgraduate level. This framework holds direct implications for curriculum design, supervisory practices, and the development of broader institutional policy.
Rethinking assessment evidence in higher education
Kristofina Junias1
Generative artificial intelligence (GenAI) has intensified concerns about academic integrity and the credibility of assessment in higher education. Institutional responses have focused heavily on detecting AI-generated work and regulating its use. This conceptual paper argues that detection addresses only part of the assessment problem because establishing whether AI contributed to a submission does not establish what the student knows or can do. Drawing on validity theory and principles of authentic assessment, the paper distinguishes academic integrity and authorship from the validity and authenticity of assessment evidence. It argues that AI does not automatically invalidate assessment evidence; its significance depends on the intended competence, assessment conditions, and whether AI supports or substitutes for the knowledge, reasoning, skills or judgment being assessed. The paper proposes an evidence-centred conceptual model that begins with intended competence, determines the appropriate role of AI, and evaluates whether student performance generates sufficient evidence to support the interpretations and decisions made from it. It further advocates multiple sources of evidence, visibility of construct-relevant processes, assessment of AI engagement where appropriate, and alignment between assessment conditions and intended claims. The paper concludes that higher education should shift from asking whether students used AI to whether assessment evidence supports defensible judgments about competence.
Between cognitive support and technological dependency
Hafssa Bourfed1; Hamza Boujida2
Something shifted in late 2022. Since ChatGPT became publicly available, Moroccan university students have folded generative AI tools into their writing practices faster than institutions have managed to respond. This paper examines how these tools are reshaping academic writing in Moroccan higher education, drawing on Actor-Network Theory (Latour, 2005) and cognitive load theory (Sweller, 1988) to read student-AI interaction as a hybrid composition process rather than a simple outsourcing of intellect. The study combines a review of recent literature on AI-assisted writing with observations from teaching practice at ENS Fes, where the question of academic integrity has moved from "did the student write this?" to "what cognitive work is still the student's own?" The usual opposition between AI as learning aid and AI as cheating misses what is happening on the ground. What we observe instead is a redistribution of cognitive load across human and non-human actors, with uneven effects depending on the student's prior literacy, language of instruction, and institutional context. Three conclusions follow. AI tools tend to reduce extraneous cognitive load but often bypass germane load, the effort that builds writing competence. Multilingual students working across Arabic, French, and Darija negotiate AI outputs differently from monolingual users. And Moroccan universities still lack a clear pedagogical framework for this transition, leaving teachers to improvise. The paper closes with recommendations for AI literacy adapted to the Moroccan context.
An AI-assisted ESP teaching experience
Wahiba Benmezal1
Teaching scientific writing to students in highly specialized STEM fields presents a challenge for English for Specific Purposes (ESP) instructors when they are not specialists in their students’ disciplines. This mixed-methods case study examined the perceived effectiveness of an Advanced Technical Writing course that used ChatGPT as pedagogical support for final-year Licence (Bachelor’s) students in Artificial Intelligence and Cybersecurity at Numidia Institute of Technology (NIT), an Algerian institute specializing in these disciplines. ChatGPT was used to develop discipline-relevant instructional materials, including examples, writing activities, and classroom exercises. Data were collected through a questionnaire completed by eleven students and semi-structured interviews with six faculty members (three supervisors and three jury members). Students reported perceived improvement in three areas of scientific writing: report organization, academic language, and preparation for final-year project writing. Faculty members also reported that the course allowed supervisors to focus on technical guidance. However, while students reported improvement in structuring scientific reports, faculty members identified difficulties in students’ final-year project reports, particularly in literature synthesis, critical comparison of studies, identifying research gaps, and methodological justification. Faculty members expressed concerns about students’ over-reliance on AI tools and its potential impact on their engagement with deeper research thinking. The findings suggest that scientific-writing instruction should combine guided practice, engagement with research articles, explicit teaching of research thinking, responsible AI use, and closer collaboration between ESP instructors and disciplinary supervisors. AI-assisted course design can support ESP teachers in specialized disciplines while highlighting the need to teach research thinking alongside scientific writing.
Integrity, authenticity, and fairness
Aissani Nouseiba1; Himoura Kawther2
The increasing use of Artificial Intelligence (AI) in education has raised significant concerns regarding assessment integrity, academic misconduct, and the need for "AI-proof" evaluation methods. In English as a Foreign Language (EFL) contexts, assessing speaking proficiency presents both challenges and opportunities for ensuring authentic measurement of learners' communicative abilities. This study explores how speaking assessment can contribute to maintaining academic integrity by emphasizing real-time language production, spontaneous interaction, and communicative performance, which are less susceptible to unauthorized AI assistance than traditional written assessments. The research investigates EFL teachers' perceptions of the use of AI-based assessments in promoting fair and trustworthy evaluation practices. A quantitative research design is adopted, using a questionnaire administered to EFL teachers working in higher education institutions. The questionnaire seeks to identify teachers' views on the impact of AI on language assessment, the risks of academic misconduct, and the potential of oral assessment tasks such as interviews, presentations, discussions, and collaborative activities to serve as more authentic and reliable evaluation tools. The collected data will be analysed descriptively to determine prevailing attitudes and assessment practices. The study aims to provide insights into the development of assessment strategies that preserve academic integrity while accurately evaluating speaking proficiency. The findings are expected to support educators in designing ethical, valid, and AI-friendly assessment approaches for contemporary language learning environments.
AI-assisted academic assessment in practice — evidence from two european pilots
Laura Dragos-Radoi
The question of how artificial intelligence can support academic assessment without displacing the evaluator has moved from theoretical debate to operational reality faster than most institutions have had time to think carefully about it. This paper presents evidence from two empirical pilots of MeritMinds, an AI-assisted academic assessment platform developed and tested in European higher education contexts, to examine what responsible AI integration in assessment actually looks like in practice. The first pilot involved 20 bachelor's theses submitted to a Faculty of Modern Applied Languages at a Romanian university (cohort mean score: 58.5%). The second involved collaboration with an independent lecturer at a European institution. Across both pilots, the platform supported evaluator judgment through rubric formalization, criterion-level consistency checking, and cohort-level pattern detection — not to generate grades or replace evaluator decision-making. The paper draws on these pilots to address three questions institutions rarely ask before adopting AI assessment tools: what does the system actually decide, and what does the human evaluator decide? How is accountability distributed when AI and human judgment diverge? And what does a partial failure — one pilot produced results requiring diagnostic rather than operational interpretation — reveal about the conditions for responsible deployment? The paper argues that these questions are not technical but pedagogical and institutional, and that answering them honestly is a precondition for any meaningful framework for AI-assisted assessment in higher education.
Mapping the gap between principles and practice
Laura Dragos-Radoi1
The past three years have produced an accumulation of frameworks, guidelines, and regulatory instruments intended to govern the use of artificial intelligence in education. The EU AI Act, institutional AI policies, and national strategies have collectively established a vocabulary for responsible AI in educational settings. What they have not yet produced is a practical account of what responsible AI adoption actually requires from institutions navigating these frameworks under real operational conditions. This paper draws on experience in AI ethics research, strategic consulting, and the co-development of an AI-assisted academic assessment platform (MeritMinds) to examine the gap between responsible AI principles as written and responsible AI as practiced. The analysis focuses on three structural mismatches that current frameworks consistently underestimate: the gap between risk classification and institutional capacity to act on it; the gap between transparency requirements and the ability of faculty and administrators to meaningfully interpret AI outputs; and the gap between accountability language in policy documents and actual lines of responsibility when AI-assisted decisions cause harm. The paper proposes a practitioner-oriented diagnostic for educational institutions, structured around three levels of readiness — awareness, integration, and accountability — and offers specific questions institutions should be able to answer before deploying AI tools in high-stakes contexts. Particular attention is given to assessment and feedback tools, which the EU AI Act classifies as high-risk applications, and to the position of higher education institutions as simultaneously AI adopters, AI developers, and AI governance actors.
Teaching students to identify bias in AI-generated writing through the VERIFY framework
Marina Jovic1
As generative AI becomes a routine part of undergraduate writing, students need instruction that moves beyond factual checking towards identifying and revising the assumptions embedded in fluent prose. This paper develops a pedagogy of bias-focused revision within the VERIFY framework: Verify Sources, Evaluate Logical Flow, Review Rhetorical Strategies, Identify Biases and Assumptions, Formulate Feedback, and Your Contribution. It draws on empirical research on cultural, demographic, and representational bias and uses a cognitive-linguistic comparison of fifteen matched essay pairs as a bounded classroom case rather than as evidence of the general prevalence of bias. The pairs consist of ChatGPT-generated argumentative essays and student revisions by Gulf Arabic-speaking EFL undergraduates responding to whether parents should influence their children's career choices. In this corpus, students did not simply copy or correct AI-generated prose; their revisions often altered how authority, autonomy, responsibility, and cultural normality were framed. On this basis, the paper proposes an initial contrastive modelling task followed by a five-step routine: mark exact wording, explain its framing effect, plan a targeted change, revise independently, and reflect on the decision. The routine is designed for two 50-minute sessions, can be compressed after students know it, and is supplemented by selective extensions rather than a fixed list of activities. The paper also offers an analytic rubric that links textual evidence, diagnosis, revision, contextual judgement, and documented decision-making. It argues that critical AI literacy should be assessed not only through students' ability to detect unreliable content, but also through their ability to justify, revise, and take responsibility for the values their final texts reproduce.
Shilpa Wadhwa1; P.Sivakumar2
Generative Artificial Intelligence (GenAI) is increasingly embedded in management education, creating simultaneous opportunities for cognitive support and concerns about overreliance. This study examines GenAI in relation to critical thinking, learning ability, and managerial decision-making among management students, while also examining faculty perspectives on AI-supported pedagogy and assessment. An exploratory mixed-method design was adopted. Data were obtained from management students through a structured questionnaire and faculty interviews. Student data were analysed using descriptive statistics, principal component analysis (PCA), and three multiple regression models using SPSS. Ninety percent of students reported using GenAI several times a week or more. PCA revealed distinct patterns of active critical engagement and passive acceptance, while learning and decision-making items produced a common AI overreliance. GenAI usage significantly predicted learning ability (β = .617, p < .001; R² = .385), critical thinking (β = .586, p < .001; R² = .381), and managerial decision-making (β = .522, p < .001; R² = .303), controlling for age and gender. Faculty interviews reinforced the duality: GenAI was valued for preparatory and efficiency-oriented tasks but treated more cautiously in assessment. The findings indicate that GenAI is neither inherently honing nor harming critical thinking; its educational value depends on verification, reflection, self-regulation, and independent judgment. This paper explores when and how generative AI hones students' critical thinking by scaffolding reasoning, diversifying perspectives, and enabling iterative feedback—and when it harms those skills—by encouraging shortcuts, cognitive offloading, superficial synthesis, and academic misconduct in management education. The study will cover AI-supported teaching design, assessment integrity, AI literacy, ethics, and critical thinking in digital environments. The study proposes a mixed-methods design combining a survey of management students and faculty with semi-structured interviews. The study offers a double-edged educational tool and proposes strategies for its responsible integration in management education.
Innovation, integrity, and the future of postgraduate scholarship in higher education
Vusi S. Mncube1
Artificial Intelligence (AI) is rapidly transforming postgraduate academic research across higher education institutions worldwide. From AI-assisted literature reviews and automated data analysis to generative writing systems and research design support, AI technologies are reshaping how knowledge is produced and disseminated. While these technologies create opportunities for enhanced research efficiency, methodological innovation, and broader access to academic support tools, they also raise significant ethical, pedagogical, and institutional concerns. This paper critically examines the dual role of AI in postgraduate scholarship by exploring tensions between innovation and academic integrity, technological enhancement and cognitive dependency, as well as accessibility and inequality within higher education systems. Particular attention is given to the implications of AI-assisted research for scholarly identity, critical thinking, originality, authorship, and methodological rigour. The paper further evaluates how unequal access to advanced AI systems may reinforce disparities between well-resourced and under-resourced institutions, particularly within Global South contexts. Drawing on conceptual analysis and emerging debates in higher education governance, the study argues for a balanced and ethically grounded approach to AI integration in academic research. The paper concludes by proposing a framework for responsible AI adoption that supports innovation while safeguarding academic integrity and critical inquiry in postgraduate education.
PRACTICES, PERCEPTIONS AND CHALLENGES. A Case Study of High Teacher Training School of Bouzareah, Algiers
Wassila Tebaa1
Currently, Artificial Intelligence (AI) and automated feedback tools are increasingly gaining significant attention, particularly in higher education EFL (English as a Foreign Language) academic writing. Despite integrating these tools into learning, little research has been conducted to examine how pre-service teachers engage with these tools during high-stakes academic writing tasks such as dissertations. This study, therefore, explores the use of automated feedback tools in EFL pre-service teachers’ dissertation writing, focusing on their actual practices, perceptions, and the challenges they face. To meet this objective, the study adopted a mixed-methods design, whereby a questionnaire was administered to 52 EFL pre-service teachers to collect both quantitative and qualitative data. The questionnaire examined patterns of use, perceived usefulness, and difficulties associated with automated feedback tools. The study indicated that automated feedback tools were inevitable in pre-service teachers’ dissertation writing, and tools such as Grammarly, plagiarism checker, and AI writing assistant were widely adopted for drafting, editing, and proofreading. The study also disclosed that these tools provided substantial assistance in enhancing writing quality and language accuracy, although participants raised some concerns related to unreliable feedback and ethical issues, including plagiarism, excessive reliance, and responsible AI use. The findings contribute to a deeper understanding of digital writing practices in EFL teacher education and offer a balanced approach integrating automated feedback tools alongside traditional supervisors’ feedback in dissertation writing, maintaining clear ethical boundaries to preserve academic integrity.
An empirical examination of AI reliance, cognitive offloading, and the homework–exam performance gap among competitive-exam candidates: evidence from a 100-students of competitive-exam preparation sector
Rohit Kumar Pal1; Sagnika Pal2
Generative AI can produce polished written responses with little effort, creating a distinction between assisted performance and competence. This study examines whether frequent AI use for descriptive homework is associated with a widening gap between homework performance and supervised, offline examination performance among competitive-exam candidates. The dataset contains 100 student records classified into low, moderate, and heavy AI-use groups, together with homework and exam scores, AI dependence, awareness, and faculty satisfaction. Homework performance rises from 62.15% in the low-use group to 73.41% in the moderate-use group and 91.21% in the heavy-use group, whereas supervised exam performance falls from 62.08% to 41.41% and then to 5.86%. Very large group differences are observed; because exam-score variances violate homogeneity, Welch's ANOVA is also reported. AI-use frequency correlates positively with homework performance (r = .938) but negatively with exam performance (r = −.955), while perceived AI dependence correlates negatively with exam performance (r = −.956). The findings are interpreted through cognitive offloading, retrieval practice, metacognition, automation bias, and self-regulated learning. The pattern is consistent with a mechanism in which AI substitutes for generative and retrieval work that normally develops independent writing skill, producing an illusion of competence when assisted output is mistaken for personal mastery. However, the source file describes the dataset as synthetic and the design is observational; therefore, the results do not constitute causal experimental proof that AI decreases human cognitive ability. They provide an empirical demonstration of a performance-assistance gap and a framework for randomized longitudinal research.
Costin-Valentin Oancea1; Diana Hornoiu1
This paper discusses the evolving relationship between Artificial Intelligence (AI), ethics, and academic integrity in contemporary linguistic research. As AI-driven technologies increasingly influence data collection, corpus analysis, translation and discourse studies, they offer extraordinary opportunities for efficiency, accuracy, and interdisciplinary innovation. At the same time, the growing reliance on AI raises important ethical concerns regarding authorship, transparency, bias, data privacy, and the responsible use of automated tools in scholarly work. This study highlights that academic integrity remains a fundamental principle in linguistic research, requiring researchers to ensure originality, proper attribution, methodological transparency, and critical engagement with AI-generated outputs. Rather than replacing human expertise, AI should be understood as a complementary instrument that supports linguistic inquiry while preserving scholarly responsibility and ethical standards. The paper argues for the development of clear ethical guidelines and responsible research practices that balance technological advancement with the core values of academic scholarship.
Atul Bamrara1
This paper proposes an AI-enabled learning analytics framework for real-time student progress tracking and adaptive educational intervention. The model integrates machine learning algorithms, predictive analytics, intelligent dashboards, and competency-based assessment mechanisms to monitor learner engagement, academic performance, cognitive progression, and behavioral indicators within digital learning ecosystems. The proposed framework employs data mining, adaptive feedback systems, and AI-assisted learner profiling to identify learning gaps and generate personalized instructional pathways. The architecture includes five interconnected modules: data acquisition, analytics engine, learner modeling, predictive intervention, and performance visualization. The model aligns with competency-based education, personalized learning environments, and NEP 2020 driven digital transformation initiatives. The framework aims to support evidence-based pedagogical decision-making, enhance learner retention, optimize instructional strategies, and facilitate continuous formative assessment. The study also addresses critical concerns related to algorithmic transparency, data privacy, ethical AI governance, and digital inclusion in educational technology systems.
Navigating artificial intelligence and academic integrity
Naimi Amara1; Yaqot Elbechir2
Artificial Intelligence is increasingly shaping how students approach academic work, yet its implications for integrity and authenticity remain troubling. This study examines the role of AI in thesis writing among ten Master's students at Chlef University, Algeria, using a qualitative approach grounded in observation and educator insight. Findings reveal a sharp contrast between the polished, fluent quality of students' theses and the weaker, less coherent writing evident in their exam responses. In the absence of institutional mechanisms to detect AI-generated text, instructors must rely on assumptions and personal familiarity with their students, a practice that undermines assessment reliability and erodes confidence in academic outcomes. Results point to a significant dependence on AI tools, raising urgent concerns about fairness, credibility, and the preparedness of higher education systems to safeguard scholarly standards. The study highlights the need for localized detection frameworks, professional training in AI literacy, and transparent dialogue among students, supervisors, and policymakers. Situated within the broader debate on AI and educational quality, this research calls for proactive strategies to ensure that technological innovation supports rather than compromises academic integrity.
Rethinking assessment strategies in higher education
Finita G Roy1
The fast development of generative artificial intelligence poses several issues with respect to the traditional approaches to the assessment and integrity of academic writing. The capabilities provided by tools such as ChatGPT and other large language models allow students to easily create papers, summaries, and research works in minutes, which makes it difficult to evaluate their authorship and originality. For this reason, many institutions doubt whether plagiarism checkers such as Turnitin can provide an accurate assessment since these technologies cannot easily recognize AI-written texts. This paper explores how artificial intelligence changes the current state of affairs in terms of academic integrity and discusses the pedagogical implications of using detection-based technologies in assessment. Through analyzing current literature on the topic and institutional guidelines regarding the problem, the paper highlights the inefficiency of these technologies and suggests developing a new approach to assessment in order to cope with challenges that arise in the age of generative AI. Special emphasis is placed on the development of other forms of assessment that are less susceptible to potential abuse by students, such as oral exams, project work, reflection tasks, and controlled writing exercises in class. This method not only lowers the risk of cheating using artificial intelligence but also fosters more profound learning and involvement from the students. Moreover, the paper highlights the need for creating AI literacy among the learners and teachers so that they can harness the benefits of AI without compromising their integrity.
The evolving role of generative AI in Indian students opting for international transfer programs (master's abroad)
Supriya Lamba Sahdev1; Chitra2; Ghousia Khatoon3; Monika Sharma4
Purpose: This study investigates the evolving role of generative artificial intelligence (GenAI) in the decision-making journey of Indian students who are researching, evaluating, and applying for international master's programs through International Transfer Programs (ITPs). Moving beyond simple adoption metrics, the study examines how GenAI usage transitions from initial curiosity-driven exploration to instrumental reliance for programme research, university shortlisting, scholarship identification, and application strategy and the attendant sentiments, risks, and information quality concerns that emerge across this trajectory. Design/Methodology/Approach: A mixed-methods approach is adopted. Qualitative data are generated through semi-structured interviews with 28 Indian students who used GenAI tools during their ITP application process. Concurrently, thematic analysis of 1,240 Reddit and YouTube discourse segments from Indian student communities is conducted using Leximancer's Topic Guide software. Findings are interpreted through the Technology Acceptance Model (TAM) extended with information quality, trust, and risk constructs. Findings: Three evolutionary phases of GenAI usage are identified: (1) an admiration phase characterized by enthusiasm for GenAI's ability to synthesize complex programme information and personalize shortlisting recommendations; (2) an instrumental phase in which students deploy GenAI for Statement of Purpose (SoP) drafting, scholarship research, and visa documentation guidance; and (3) an apprehension phase marked by emerging concerns over hallucinated programme data, outdated fee structures, ethical ambiguity in SoP co-creation, and reduced engagement with official university counsellors. Social influence from peer networks significantly mediates GenAI adoption intensity, while perceived information risk moderates the transition from adoption to deep usage. Research Limitations/Implications: The study is confined to Indian students targeting English-speaking destination countries. Longitudinal tracking of actual enrolment outcomes via GenAI-informed decisions is a critical future direction. Practical Implications: International offices, education consultants, and destination universities should develop GenAI-aware advising frameworks that complement rather than compete with students' AI-mediated research behaviour. Indian HEIs offering ITP pathways should leverage GenAI's counselling potential while mitigating misinformation risks through verified AI content pipelines. Originality/Value: This is among the first studies to examine the full arc of GenAI usage from adoption to deep decisional reliance in the specific, high-stakes context of international student mobility and transfer program selection in India. It offers a novel three-phase GenAI usage evolution model grounded in TAM and information quality theory, with direct implications for international education advising practice.
Yoke Lian Lau1
The rapid integration of generative artificial intelligence (AI) in our everyday lives is creating an 'academic double standard' in many higher education institutions where faculty members are being encouraged to utilize AI in their work environment to increase productivity however, students will be restricted when it comes to assessments of their knowledge. This study examines the tensions existing in the context of classroom-based language assessment, particularly on the topic of Mandarin as a foreign language (MFL) composition writing assignments at the Universiti Malaysia Sabah (UMS). A mixed methodology applies both quantitative surveys and qualitative interviews were used to explore the views of two populations: 23 Mandarin language students and 7 MFL instructors regarding the usefulness of AI, ethical issues related to institutional uses of AI, and how AI affects teaching practices. The results of the quantitative survey indicate that students view AI through a practical perspective; 69.6% see AI primarily as an instructional tool allowing them to actively negotiate linguistic complexity prior to submitting written compositions instead of viewing AI as a means to cheat. Both student respondents and instructor respondents agree that AI provides high value, instantaneously assists learners with logographic and tonal complexities inherent to Mandarin Chinese. Nevertheless, each of the seven instructors surveyed agreed that if learners rely too heavily upon AI produced written output they may ultimately mask a learner's true ability to communicate. Additionally, this research identifies a significant gap in the governance of these technologies within higher education: While 43.5% of students clearly identify an institutional double standard as perceived by themselves as unfair, 100% of the instructor respondents indicated that they have utilized AI professionally and never received formal instruction or training on ethical considerations related to AI. Therefore, this paper concludes that blanket prohibitions on technology should not exist; Rather, we advocate that higher education institutions develop consistent and transparent guidelines and policies governing the use of AI within educational settings, and shift towards developing process oriented, multimodal assessment frameworks such as oral vivas and drafting portfolios to assess true language proficiency.
Perceived benefits, limitations, and impact on research originality: a case of first-year master's students (M1), applied linguistics — Skikda university academic year 2025–2026
Ahlem Boumediene1
The swift increasе in the use оf аrtificial intelligenсe (AI) tооls within higher educatiоn has sparked signifiсant aсаdemic disсussiоns abоut their functiоn in rеsearch, еspeciаlly аmоng pоstgraduаte students. This reseаrch ехplоres thе pеrspectives оf first-yеar Mаstеr's studеnts in Applied Linguistics at Skikdа University regаrding thе аdvantages аnd drawbacks оf AI tооls whеn sеlесting dissertаtiоn tоpiсs, as well as thе pоtential cоnsequenсes fоr the оriginality оf thеir rеsearch. An online quеstiоnnaire was distributed to 41 students using а miхed-methоd approach during the 2025-2026 acadеmiс year.The study intеgrаtes frеquеnсy аnalysis оf сlоsed-endеd questiоns with themаtiс cоding оf оpen-еndеd rеspоnses. The outcomes show that over a third of students (46.3%) used the AI tools to generate ideas, and just under half (39%) used them to guide their literature review, rather than for detailed analysis of their work .A substаntial pоrtiоn оf thоsе survеyed, amоunting tо 53.7%, favоrеd а mоdel fоr AI integratiоn that emphаsizеs regulаtiоn and trаnsparеncy in disсlоsures. Meanwhilе, an оverwhelming 95.1% аdvоcated fоr the estаblishment оf fоrmal pоlicy mеasures within institutiоns. Tоgеther, thеse insights suggеst а mоdel fоr integratiоn thаt is cоnditiоnal and based оn funсtiоnаl соnstraints, critical mеdiatiоn, institutiоnal suppоrt struсturеs, and rеfоrms in teаching аpprоасhеs. These results hаve impоrtant cоnsequеnces fоr the gоvеrnаnсе оf AI, the praсtices surrоunding dissertаtiоn supеrvisiоn, and the dеvelоpment оf curriсula in the cоntехt оf highеr educatiоn in Algеriа.
Otmane Azeroual1
Artificial Intelligence (AI) is increasingly transforming research management processes in higher education and research organizations. From automated researcher profiling and publication analysis to funding intelligence, expert recommendation, and research evaluation, AI-driven systems are becoming integral components of modern Research Information Systems (RIS) and Current Research Information Systems (CRIS). While these technologies promise greater efficiency, scalability, and data-driven decision-making, they also raise critical questions regarding transparency, data quality, accountability, and trust. This paper examines the opportunities and challenges associated with the adoption of AI in research management and academic knowledge infrastructures. It argues that the effectiveness of AI-based research management depends not only on algorithmic performance but also on the quality, completeness, and governance of the underlying research information. Inaccurate metadata, fragmented data sources, opaque algorithms, and AI-generated misinformation can significantly affect research assessment, funding decisions, institutional reporting, and strategic planning. Building on concepts from Responsible AI, Data Governance, and Research Information Management, the paper proposes a conceptual framework for Responsible Academic Intelligence. The framework emphasizes transparency, explainability, fairness, data quality, human oversight, and ethical governance as key principles for trustworthy AI deployment in academic environments. Furthermore, the paper discusses the role of knowledge graphs, semantic technologies, and human-in-the-loop approaches in improving the reliability and accountability of AI-supported research management systems. By bridging the fields of artificial intelligence, higher education, and research information management, this contribution provides practical and strategic recommendations for universities, research organizations, policymakers, and system developers seeking to implement trustworthy AI solutions. The paper contributes to the growing discussion on how academic institutions can leverage AI responsibly while preserving scientific integrity, transparency, and trust in the age of intelligent research ecosystems.
Muhammad Mooneeb Ali1
Artificial intelligence (AI) has bottlenecked the established operational procedures by swapping them imperiously. It has also embarked its vigorous presence in academia. Contemporary research is eyeing on the generative aspect of AI with successful results. Attributing its significance, this investigation will delve into the area of AI for assessment purposes; precisely, assessment of English language learners' writing abilities in Essay writing through a purposefully developed AI app. The population for the specific enquiry will be Saudi Arabian students. The sample will be collected in two phases. Initially, the plan is to chose 100 Bachelor level students enrolled in an English compulsory course in semester 1 at 5 different universities to be the part of the investigation. For a valid data collection, a placement test is planned to be executed from all the participants. Later they will be placed into a focused group of equal ability to maintain validity of the sample. According to their course content, the learners will be trained on essay writing prompts only. After a month, a writing milestone assessment will be conducted, which will consist of an essay with 50 marks. Further,10 raters will be assigned for rating in two groups. 5 raters will assessed learners manually, whereas the other group of 5 raters will be using AI developed app as a preliminary method for assessing essays. IELTS writing rubric is planned to be used for assessment by both groups of raters. All the raters will assess 10 essays. It is presumed that results will show the raters who used AI as their preliminary assessment as more consistent, efficient, reliable, and valid in their assessment in comparison to orthodox raters (without AI app). Moreover, they will be also be providing comprehensive feedback to the learners. This research can serve as one of he benchmarks in AI into assessment area.
Hafiane Insaf1; Hafsi Latifa2
Artificial intelligence has witnessed a notable expansion recently, for being incorporated into a multitude of academic and professional aspects. This study investigates the lexical density of AI-edited medical abstracts compared with their original versions. Using a quantitative corpus-based comparative design, 30 PubMed-indexed medical article abstracts published between 2019 and 2021 were selected and edited using a standardized prompt in ChatGPT-5. Lexical density was measured with AntConc following Ure's (1971) framework. The findings revealed highly similar lexical density levels between the two datasets. Human-written abstracts obtained a mean lexical density score of 62.06 (SD = 3.83), whereas AI-edited abstracts recorded a mean score of 62.10 (SD = 3.12). Although the AI-edited versions displayed slightly greater consistency in lexical distribution, the difference in lexical density was statistically negligible. The results indicate that AI-assisted editing has only a limited effect on lexical density and preserves the informational load and overall lexical characteristics of medical article abstracts.
Opportunities, ethical challenges, and future directions in scientific research
Abdelkader Benammar1; Amina Bendaoudi2; Zoubir Mahdjoub2
The swift incorporation of artificial intelligence (AI) into scientific inquiry has transformed how researchers collect data, analyze it, generate text, and communicate their findings. AI-based tools offer significant opportunities to improve research efficiency, support academic writing, facilitate literature exploration, and accelerate knowledge production. However, the widespread adoption of AI also raises important concerns regarding academic integrity, authorship, transparency, originality, and ethical responsibility. This paper explores the dual role of artificial intelligence in scientific research by examining both its benefits and associated risks. The study discusses how AI technologies can support researchers while also creating challenges related to plagiarism, inappropriate authorship attribution, bias amplification, and overreliance on automated content generation. Particular attention is given to emerging academic policies and ethical frameworks intended to guide responsible AI use in higher education and research environments. The study argues that maintaining academic integrity in the era of AI requires a balanced strategy that integrates technological advancement with ethical awareness, institutional oversight, and transparent research methodologies. Proposals are put up to promote the responsible integration of AI, ensuring trust, originality, and scientific integrity.
Abir Bendebbache1
The growing use of generative AI in EFL writing assessment has raised interest in the reliability and validity of AI-generated evaluation, particularly its alignment with human teachers' contextual and critical judgment. By translating writing quality into explicit criteria, analytic rubrics may support more transparent and consistent assessment. This study investigates whether rubric-guided prompting produces ChatGPT-generated scores closer to human judgment than unguided AI scoring. It analyzes 18 responses by second-year EFL students from the University of Oum El Bouaghi, Algeria. Each response, comprising a source-text summary and argumentation, was assessed under three conditions: unguided ChatGPT-5.5 scoring, rubric-guided ChatGPT-5.5 scoring, and human scoring using the same adapted rubric. Based on Yamanishi et al.'s L2 summary-writing framework and Toulmin-informed argumentative-writing assessment, the rubric assessed content understanding, paraphrase quantity, paraphrase quality, claim/viewpoint, argument development, and writing quality. Each criterion was scored out of 4 points, with a 24-point total later converted to 20, while unguided AI generated holistic scores out of 20. The findings challenge the assumption that rubric-guided prompting aligns AI scoring with human judgment. Unguided AI was closer to human scores in 14 of 18 texts, with a mean absolute difference of 1.24 points, compared with 2.32 for rubric-guided AI, which tended to score more generously. The study concludes that rubrics improve transparency and structure AI feedback, but do not ensure pedagogical calibration, supporting AI as a formative assistant rather than an autonomous evaluator. Responsible AI integration therefore requires teacher training in AI literacy, prompt design, and critical monitoring.
Bob Barrett1; Linda C. Ashar2
Educators have been challenged with assessing learning since learning was first conceived as a discipline. From muscle memory to harnessing imagination, learning is how humans survive and progress individually and as a species. Artificial intelligence (AI) dramatically proves the very things humans learn to do challenges how they will continue to do it. This process expands enlightenment on how we think and learn. AI's rapid integration into education along with every other facet of modern life, has transformed the methodologies of teaching, researching, and completing projects and assignments. While AI-powered systems and tools offer opportunities for personalized learning and increased efficiency, they also create significant challenges for those responsible for assessing student achievement. This essay examines the impact of AI on traditional assessment practices and the corresponding issues of academic integrity and the meaning of authentic leaning. How can educators effectively assess student learning in an era where AI tools can generate written work, solve problems, and assist with academic tasks? This paper argues that interactive, participative, and collaborative assessment techniques can demonstrate successful learning while maintaining academic rigor.
Students' perceptions and experiences
Moumeni Abderrahim1; Mouili Fatiha1
The rapid spread of ChatGPT has changed the way university students practice academic writing. While many students use it to generate ideas, organize their papers, improve vocabulary, and correct grammar, there are increasing concerns about over-reliance on the tool, which may negatively affect their critical thinking and personal writing style. This study aims to explore university students' perceptions of using ChatGPT in academic writing. The research will follow a quantitative approach by collecting data through an online questionnaire distributed to 50 university students. The survey will examine how frequently students use ChatGPT, the specific writing tasks they use it for, its perceived advantages and disadvantages, and its effect on the development of their writing skills. The findings are expected to help educators understand how to integrate ChatGPT effectively and responsibly into academic writing courses.
Artificial intelligence ethics awareness and ethical usage among preservice teachers in universities in Ogun state
Akorede Ayoola Asanre1; Adeleke, Heritage Oluwadiwamu2
The rapid adoption of Artificial Intelligence (AI) in higher education has reshaped how preservice teachers learn, prepare, and engage with academic tasks. Despite its benefits, concerns about ethical usage, such as academic integrity, transparency, and responsible engagement, continue to grow. A key issue is whether awareness of AI ethics translates into ethical practices among preservice teachers. This study investigates the relationship between AI ethics awareness and ethical usage in two selected universities. Using a quantitative predictive correlational design, the study involves preservice teachers across all academic levels selected through convenience sampling. Data were gathered using a structured 5-point Likert-scale questionnaire to get information on preservice teachers' level of AI ethics awareness and Ethical usage. It was validated using both face and content validity. The reliability coefficient gave 0.78 using Cronbach alpha. Data analysis was done using descriptive statistics, Pearson correlation, and simple linear regression to determine levels, relationships, and predictive influence among variables. The study is expected to provide insight into whether awareness of AI ethics is sufficient to promote responsible AI use among future teachers and to highlight implications for strengthening ethical AI integration within teacher education programmes and institutional policy frameworks.
Ayah Dadi1; Nassira Houat2
The rapid integration of generative artificial intelligence tools into academic life has forced universities worldwide to ask an uncomfortable question: when students use AI, are they cheating, or simply adapting to the tools of their time? In Moroccan higher education, this tension is particularly visible, as institutions struggle to enforce academic integrity policies that were designed long before tools like ChatGPT existed. This study aims to examine how students across different disciplines perceive and justify their use of artificial intelligence in academic work, and to interrogate whether existing misconduct policies adequately reflect the realities of today's digital learning environment. Drawing on semi-structured interviews conducted with 53 students from multiple faculties across Moroccan universities, including humanities, sciences, and engineering, this research captures a diverse range of student voices, motivations, and ethical reasoning. Findings reveal that the majority of students do not perceive AI use as dishonest, but rather as a practical and necessary academic survival strategy, particularly in contexts where guidance from institutions remains absent or contradictory. Students consistently highlighted the gap between what policies prohibit and what their academic workload practically demands. These results suggest that punitive approaches to AI use are not only ineffective but risk deepening student disengagement. Universities are therefore urged to move beyond policing toward policy reimagination, developing clear, contextually sensitive, and educationally grounded frameworks that treat artificial intelligence as a pedagogical reality rather than an institutional threat.
A process-oriented scaffolding framework for second-language higher education
Seemab Jamil Ghouri1
The rapid development of generative artificial intelligence (GenAI) as a disruptive agent in the modern period offers postsecondary education both previously unheard-of potential and significant limitations for assessment design. Many traditional exams are now in danger of becoming outdated because GenAI can produce credible assessment answers with limited student involvement. Discussions prevailing sector-wide are around how to redesign assessment to make it possible for the appropriate and ethical use of GenAI whilst ensuring students actually achieve the learning outcomes expected. Quality assurance bodies have called for updated strategies but there is a large disparity between the speed of GenAI progress and the implementation of such assessment reforms. Therefore, the Purpose of this paper is to examine how university faculty members can create tests which could efficiently assess the undergraduate performance amid the availability of GenAI at the disposal of all students. Using a mixed methods design of empirical research in Pakistani universities, this study employs a multi-stage approach of assessment scaffolding wherein emphasis in grading changes from the final product to the continuous process of learning itself. This is done by gathering both quantitative as well as qualitative feedback on such an approach. This paper provides an empirical base for curriculum developers seeking to develop an assessment system that would preserve bilingual integrity without compromising on the development of writing skills.
Minkwon Kim1
Automated essay scoring (AES) is increasingly used in high-stakes educational assessment, while raising questions such as demographic fairness. A growing body of work evaluates predictive bias by comparing model scores to human scores conditional on essay quality. However, this established approach, including recent studies on the ASAP 2.0 corpus, conditions on the human rating as if it were unbiased ground truth, while human scoring carries documented, group-varying reliability. Treating it as fixed truth conflates bias the model introduces with bias inherited from a differentially reliable benchmark. This study proposes a measurement framework that re-expresses AES predictive bias relative to the human-reliability ceiling, separating model-introduced from benchmark-inherited bias. Using the open ASAP 2.0 corpus, we fine-tune encoder-based scorers (RoBERTa, DeBERTa) and estimate differential prediction across English Language Learner status and economic disadvantage. These two variables anchor the analysis because they exercise the construct-relevance distinction in opposite directions. We benchmark observed group gaps against the corpus's reported inter-rater reliability and test whether bias patterns are concordant across models. The broader contribution is diagnostic. Fairness auditing presupposes corpora carrying both demographic metadata and documented human-reliability baselines, resources most languages lack. We argue for the minimal metadata a corpus must carry to make AES fairness checkable at all, framing auditability as a precondition for equitable assessment beyond English-dominant contexts. This reframes fairness from a property one measures post hoc to one that a corpus must be built to permit.
A corpus-based study of AI-generated and human writing
Anannya Trivedi1
As artificial intelligence becomes increasingly involved in the production of academic, creative, and literary writing, the distinction between human-authored and AI-generated text has become more difficult to establish. In response, AI-detection tools have been widely adopted, particularly within educational institutions, where their results often carry significant academic consequences. However, growing evidence suggests that these systems frequently produce inconsistent outcomes and rely on statistical markers whose reliability remains questionable. This study investigates three such markers commonly associated with AI detection namely repetition, Shannon entropy, and variation in sentence length to examine whether they can meaningfully distinguish AI-generated writing from human literary text. The proposed study compares two texts: a human-written abstract of George Orwell's Animal Farm and an AI-generated text produced from a prompt based on the same narrative. This comparison allows for an examination of whether statistical markers commonly associated with AI writing can reliably distinguish between the two. Using Python-based computational analysis, the study measures patterns of repetition, lexical predictability, and sentence-length variation while interpreting the findings through literary and stylistic perspectives. Consequently, the study argues that statistical markers cannot be treated as independent or definitive evidence of AI authorship. By bringing literary stylistics into conversation with computational approaches, the study aims to challenge the assumption that AI-generated and human-written language exist as clearly separable categories. Instead, it proposes to demonstrate that significant stylistic overlap exists between the two, raising important questions about authorship and the reliability of some common AI-detection systems.
Metacognitive reorientation and learner autonomy in AI-assisted EFL writing
Nawal Khelalfa1
This qualitative exploratory multiple-case study investigates how relatively autonomous Algerian EFL undergraduate learners negotiate metacognitive regulation, cognitive offloading, and learner autonomy in AI-assisted and non-AI writing tasks. Nine participants completed comparable writing tasks under both conditions and recorded structured reflections before, during, and after writing. Data were analyzed thematically using a theory-informed approach drawing on metacognitive regulation, cognitive offloading, and learner autonomy frameworks. Contrary to concerns that generative AI promotes passive dependence or "metacognitive laziness," participants demonstrated sustained and often expanded metacognitive engagement when using AI. Rather than reducing regulation, AI prompted a reorientation of metacognitive activity toward monitoring AI reliability, evaluating suggestions, preserving intended meaning, and managing boundaries of delegation. Cognitive offloading remained selective, with higher-order processes such as ideation, argumentation, creativity, personal judgment, and authorial voice consistently protected from delegation. Cross-case analysis identified three learner orientations—ideological resisters, strategic minimal offloaders, and strategic conditional offloaders—all characterized by strong autonomy and active regulation of AI involvement. The findings support a cognitive offload paradox whereby delegating certain writing functions reduces some composing demands while simultaneously increasing requirements for evaluation, verification, and control. The study contributes to emerging theorization of AI-assisted writing through the concepts of metacognitive reorientation, dual-layered metacognitive regulation, and protected cognitive domains.
Tackling AI-assisted dishonesty in EFL online assessments at Chadli Bendjedid University.
Ladaci Naima1
In today's academic landscape, the transition to online education has become a non-negotiable aspect to align with the international academic standards. However, the shift to online teaching and assessment in EFL contexts at El Tarf University in Algeria does not only stem from the need to align with modernization but also as a practical solution to challenges such as large class sizes and limited faculty resources. Paradoxically, this solution became a source of concern in itself. Teachers constantly report a persistent and a huge disparity between students' online results and their face-to-face performance, with online results surpassing those recorded in classroom contexts, which is more likely due to their reliance on AI tools during online exams. This paper discusses the issue of AI-assisted dishonesty in online assessments that threatens academic integrity and evaluation credibility in EFL contexts at the aforementioned university. It attempts to understand students' behaviors in online assessments by identifying the reasons behind those behaviors and then provides some practical strategies to detect and prevent them. To gather insights for this study, a mixed methods approach combining teachers' interviews, students' surveys and analysis of online assessment' responses was adopted. The findings of the study intend to assist fellow teachers and administrators in designing solid assessment practices and in implementing mechanisms that do not only uphold integrity but also provide accurate reflections of students' competence. Ultimately, the study aims to foster both the responsible uses of AI tools in EFL academic contexts and the ethical digital literacy more broadly.
Accuracy feedback and error correction
Sila Arici1
This study introduces an ongoing classroom-based teaching practice that integrates AI tools into the teaching of academic paragraphs and essays. The practice is implemented with preparatory year students—first-year university students who complete an intensive English program before beginning their departmental studies. In this approach, AI tools are used at the revision stage to provide targeted feedback on cohesion, grammar and vocabulary, while instructor feedback remains central to meaning, coherence, and idea development. The study demonstrates how AI-supported accuracy feedback can complement instructor guidance without replacing the process-based nature of academic writing instruction.
Evidence-based frameworks for learning analytics, student success prediction, and ethical implementation in higher education
Archana Pandey1
As higher education institutions globally deploy digital infrastructures, the convergence of learning analytics, predictive modeling, and automated psychometrics is reshaping academic evaluation and institutional governance. This study examines machine learning architectures designed to track student engagement, forecast attrition, and automate assessments, evaluating their ethical alignment with United Nations Sustainable Development Goal 4 (Quality Education) across diverse geopolitical regions—including Canada, the European Union, the Western Balkans, the MENA region, and the Global South. While these technologies provide scalable instructional diagnostics, they introduce acute socio-technical vulnerabilities. Left unmanaged, automated grading engines often suffer from construct drift—evaluating dominant linguistic styles over authentic mastery—thereby marginalizing non-native student populations and regional vernaculars across the Global South and Euro-Balkan corridors. Concurrently, predictive analytics can cause structural tracking, trapping vulnerable demographics in self-fulfilling negative educational loops based on historical, localized training data.
Integrating AI-Use Levels, Stakeholder Roles, and Ethical Safeguards in University Assessment
Aya Ismail
Generative artificial intelligence (GenAI) has disrupted conventional university assessment, leading many institutions to oscillate between blanket prohibition and unrestricted use. Existing models, such as the Artificial Intelligence Assessment Scale, clarify different levels of acceptable AI involvement, while policy and ethics frameworks emphasise governance, training, privacy, fairness, transparency, and accountability. However, these strands are often treated separately, leaving students and educators without practical guidance that connects how much AI may be used with who is responsible and which ethical safeguards are required. This conceptual paper proposes a human-centred Ethical Responsibility Matrix for GenAI-supported university assessment. Drawing on a structured conceptual review and framework synthesis of recent peer-reviewed literature on AI assessment, AI literacy, academic integrity, and educational governance, the study integrates three dimensions: levels of AI involvement; the responsibilities of students, educators, and institutions; and ethical safeguards appropriate to each level. The matrix differentiates between no-AI assessment, AI-assisted planning, AI-supported feedback, human–AI collaboration, and AI exploration or full integration. At each level, it specifies requirements concerning disclosure and documentation, critical verification of AI outputs, data privacy, equitable access, and final human accountability. The proposed framework shifts the debate from detecting misconduct towards designing valid, transparent, and learning-centred assessments. It is intended as an internationally adaptable decision-making tool rather than a single universal policy. By aligning permitted AI use with explicit responsibilities and safeguards, the paper seeks to preserve academic integrity and student agency while enabling the pedagogically meaningful use of GenAI in higher education.
Introducing the RAIC Model in Higher Education
Ghinwa Jalloul
Generative artificial intelligence (AI) is increasingly used in universities for teaching, assessment, research and administration, offering opportunities to improve learning and institutional efficiency while raising concerns about academic integrity, privacy, fairness and human oversight. Universities therefore need to ensure that AI is introduced without compromising educational quality and accountability. Several responsible AI frameworks provide guiding principles but putting them into practice remains a challenge. Universities need more than policies. They need the capabilities to implement responsible AI practices and assess how effectively they work. This paper proposes the Responsible AI Capability Model (RAIC). It draws on a preliminary review of existing responsible AI governance frameworks and organisational capabilities research and is informed by the ongoing development of Ethics4AI; a prototype platform designed to support the practical implementation of Responsible AI. The proposed approach maps responsible AI requirements to organisational capabilities and identifies indicators that could be used to assess how effectively these capabilities are implemented. RAIC connects four elements: governance requirements, organisational capabilities, evidence of implementation and actions for improvement. It focuses on five areas: strategic leadership, governance and accountability, human competencies, technology and data governance, and organisational learning and adaptation.
The paper uses higher education to demonstrate how RAIC could help universities assess their ability to govern AI across teaching, assessment, research and administration, identify gaps and prioritise improvements. The model remains conceptual and will require further development and empirical validation.
Exploring generative AI use in EFL student writing through classroom observation and interviews
Amira Sid1
The rapid development of generative artificial intelligence (AI) has created new challenges for English as a Foreign Language (EFL) writing instruction and assessment. This study explores how AI writing tools may influence students' writing practices, confidence, and understanding of authorship. The study emerged from the researcher's classroom observations while teaching Written Expression to second-year EFL students at Mohamed Chérif Messaadia University, Souk Ahras, Algeria. During argumentative writing activities, differences were observed between students who appeared to rely mainly on classroom-taught writing procedures and students whose writing raised questions about possible AI assistance. In particular, some students showed strong adherence to the structure taught in class despite grammatical and spelling errors, while other writing appeared more linguistically polished but did not consistently follow the required structure. To explore these observations further, the study adopts a qualitative classroom-based design combining teacher observation with semi-structured interviews with six students. The study focuses on students' experiences with AI writing tools, their reasons for using or avoiding them, their confidence in writing, and their views of authorship. The study aims to contribute to current discussions of AI in EFL writing by considering not only the quality of the final written product but also students' engagement with the writing process and the structures taught in class.
Opportunities, challenges and the evolving role of the teacher
Milica M. Kočović Pajević1
Artificial intelligence is rapidly reshaping educational practice: it is no longer just a source of information for teachers, it is becoming an active participant in how teaching materials get designed and complemented. This shift is especially visible in English for Specific Purposes (ESP), where materials must respond to learners' professional needs and disciplinary contexts. Generative AI tools now enable teachers to create customized texts, communicative tasks, case studies, and assessment activities with unprecedented speed and flexibility, thus effectively turning AI into a kind of collaborative partner in instructional design rather than a threat to the teacher's job. However, the increasing use of AI-generated materials also raises important questions regarding their pedagogical quality, linguistic accuracy, contextual relevance, ethical use, and reliability. This paper examines the opportunities and challenges associated with AI as a co-designer of ESP learning materials. Drawing on research in ESP materials development, textbook evaluation, and the growing literature on generative AI in education, it considers how AI can support lesson planning and the creation of workplace communication tasks informed by authentic professional practices, without losing sight of why teacher judgment still matters in evaluating and adapting what AI produces. The paper argues that effective integration of AI into ESP teaching depends not only on technological competence but also on teachers' pedagogical judgment and AI literacy. Rather than reducing the teacher's role, AI expands professional responsibilities by requiring educators to critically evaluate, refine, and ethically implement AI-generated materials. The paper concludes by proposing key principles for the responsible use of AI in ESP materials design and discusses implications for teacher education, institutional policy, and future research.
A pilot framework using local LLM-as-judge evaluation
Edgar Eduardo Ceh-Varela1; Yitzen Mayitza Lizama-Peraza2; Sarbagya Ratna Shakya1
This work presents an ongoing pilot design-based project that explores the use of AI-generated personas for the development and evaluation of ESL/EFL instructional materials. Inspired by multiperspectivity approaches in humanities education, the project adapts the concept of multiple perspectives to foreign language learning by generating different versions of the same English text through distinct linguistic and communicative personas, including academic, informal, persuasive, simplified learner-oriented, and culturally sensitive variants. The framework is intended to support awareness of how register, tone, vocabulary choice, perspective, and communicative intent influence meaning construction in English discourse. The project also investigates the feasibility of using locally deployed large language models (LLMs) as preliminary evaluators of AI-generated educational materials while maintaining human oversight in pedagogical decision-making. Using Ollama as a local deployment environment, the proposed framework incorporates an LLM-as-judge evaluation pipeline in which generated outputs are assessed according to a shared rubric covering register appropriateness, CEFR-level suitability, clarity, pedagogical usefulness, and visibility of perspective-related linguistic features. A subset of outputs will additionally be reviewed by a human ESL/EFL expert to provide comparative validation. Rather than reporting the results of a classroom intervention, the paper presents a scalable and low-cost methodological framework for the preliminary development and evaluation of AI-assisted ESL/EFL learning materials. The discussion focuses on the project's design rationale, evaluation methodology, and the potential benefits and limitations of combining AI-generated instructional content with automated evaluation systems in language education.
EFL students' perceptions of AI writing assistants and the risk of academic dependence
Djouadi Djawhara1; Kerroum Fathia2
The rapid integration of artificial intelligence tools into higher education has fundamentally reshaped how students approach academic writing. While AI writing assistants offer substantial support in terms of grammar correction, text generation, and structural guidance, growing concerns have emerged regarding their potential to foster dependency rather than competence. This study investigates the perceptions of Algerian EFL university students toward AI-assisted academic writing, with a particular focus on the tension between productive scaffolding and passive reliance. Drawing on questionnaire data collected from 57 undergraduate students, the study explores the frequency and nature of AI tool use, students' self-reported awareness of dependency risks, and the implications for academic integrity and autonomous writing development. The findings reveal a paradox characteristic of AI-mediated learning: students acknowledge the cognitive convenience of AI tools while simultaneously recognising their undermining effect on independent writing skills. These results are discussed within the framework of Vygotsky's Zone of Proximal Development and critical digital pedagogy, offering practical recommendations for educators seeking to integrate AI tools responsibly. The study contributes to an emerging body of research on ethical AI use in EFL contexts across the Global South.
Mohammed Fouad Rebiai1; Mouna Fethiza Tedjani2
The emergence of generative artificial intelligence (AI) tools and their evolving integration into education has reshaped the vision of language learning. AI chatbots such as ChatGpt have been seen as facilitating tools that boost the learning process as they provide learners with immediate access to explanations, feedbacks, demonstrations, and communication opportunities. While the primary focus of previous investigations on AI and language learning have been highlighting its role with teaching grammar, vocabulary, and speaking and writing skills, its potential impact on learners' pragmatic competence remains underexplored. Pragmatic competence that centers on language in context emphasizes both its appropriate use and interpretation, and it has traditionally been requiring human guidance and cultural awareness. This study investigates Algerian EFL teachers' perceptions of the contribution of AI tools in pragmatic instruction and explores the potential influence of these digital tools on teachers' professional identity. In addition, this research examines whether EFL teachers perceive AI as a supportive tool or as a threat to their position in improving students' pragmatic competence. A perception-oriented qualitative design will be adopted for this investigation. Semi-structured interviews with Algerian EFL teachers from different higher education institutions will be conducted to collect the research data, which will then be analyzed thematically. The anticipated findings suggest that participants will consider the effectiveness of AI tools in developing students' pragmatic competence through providing relevant examples and interactive scenarios while still emphasizing the role of educators in contextualizing meanings, illuminating sociocultural norms, and explaining speakers' implied intentions. This study is also expected to shed light on the impact of emerging AI tools on reshaping teacher identity beyond mere pedagogical expertise. The findings may contribute not only to teacher preparation in the digital era, but also to current debates on the ethical integration of AI into academia.
A survey-based study of Algerian university learners' practices and experiences
Aggoune Amina1; Bennacer Fouzia1
The widespread integration of digital technologies in education has reshaped English as a Foreign Language (EFL) language learning. ChatGPT has emerged as one of the most widely adopted application, due to its ability to offer immediate feedback, facilitate communication, and assist learners in their learning process. Regardless of its popularity, research remains restricted concerning the outcomes of ChatGPT on learners engagement and learning experiences in EFL contexts. Therefore, the current study aims to explore the role of ChatGPT on Algerian university learners' participation and overall learning experience at the university level. It focuses on how learners interact with ChatGPT and how this interaction may support their participation, motivation, and learning process. In response to this, the present study adopts a quantitative survey-based design involving EFL university learners in Algeria who have previous experience using ChatGPT for learning English. In addition to this, Data are collected through an online structured questionnaire based on a 5-point Likert scale to measure two main variables: learners' engagement and learning experience. The study employs descriptive and inferential statistical techniques, including means, standard deviations, correlation analysis, and regression analysis, to examine the relationships between the uses of ChatGPT, engagement, and learning experience. The findings indicate a positive relationship between learners' use of ChatGPT and their participation, motivation and learning experience. These findings also suggest that AI-assisted language learning environments can create more flexible and interactive opportunities for learning beyond the traditional classroom. Moreover, it is recommended that optimizing the integration of ChatGPT and other AI tools in education requires leveraging their opportunities while systematically addressing associated challenges. Such effective integration has the potential to advance AI-assisted language learning and enhance the overall quality of language learning.
Algorithmic mediation and compulsory language self-rescue among adult Algerian learners in CNEFD distance education
Leila Aouam1; Leila Benstaali1
Adult basic education across the Global South typically excludes foreign languages. In Algeria, the Office National de l'Alphabétisation (ONEA) delivers this education exclusively in Arabic; however, graduates entering the Centre National de l'Enseignement à Distance (CNEFD) face a concealed structural contradiction: French and English are compulsory examination subjects, despite learners having no prior foreign language instruction. This qualitative study investigates this institutional barrier, introducing the concept of compulsory language self-rescue—a condition where digitally marginalized adult learners must construct improvised, invisible learning ecologies to survive systemic exclusion. While research addresses distance learning efficacy, empirical gaps remain regarding how Arabic-medium CNEFD enrollees navigate these mandatory foreign language demands. This study examines their autonomous strategies, specifically focusing on algorithmic mediation via translation tools alongside peer networks and pattern memorization. Employing Interpretative Phenomenological Analysis (IPA) and the Critical Incident Technique (CIT), the study explores the lived experiences of 30 purposively sampled ONEA graduates across Algeria, encompassing both examination passers and dropouts. Data collected through semi-structured interviews map educational trajectories, examination preparation rituals, and decisive crises where institutional absence intersects with algorithmic survival tools. The findings expose the concealed costs of linguistic exclusion, demonstrating that self-directed and resilient behaviors emerge from structural necessity rather than pedagogical preference. Practically, the study reveals that basic literacy credentials function as a trap disguised as opportunity, offering direct implications for ONEA curriculum reform, CNEFD policy, and broader ethical dialogues surrounding algorithmic dependency in marginalized educational settings.
Opportunities, limitations, and learner perceptions
Mohammed Suhail Banqusly1
As more human interaction is exploding with AI machines, a humongous gate has opened for second language (L2) speaking practice, yet the pedagogical realities of implementing AI chatbots in actual classrooms remain underexplored. This case study investigates the integration of a customized AI chatbot into spoken English instruction for 82 male elementary and intermediate learners at a Syrian university. Drawing on Jeremy Harmer's (2015) pedagogical framework for speaking instruction and contemporary second language acquisition theory, the study examines both learner outcomes and pedagogical implications of AI chatbot use over an eight-week period. As the classroom teacher, I implemented random AI chatbots as a supplementary speaking practice tool, observing and documenting learner engagement, progress, and challenges. Findings reveal a rough 30% overall improvement in speaking proficiency, with a clear distinction between learner levels: intermediate students demonstrated some gains, communicating effectively with the AI chatbot and benefiting from independent practice. However, elementary students struggled significantly to communicate with the AI chatbot, requiring extensive teacher guidance and mediation to benefit from the tool. The study proposes pedagogical guidelines for responsible AI integration, emphasizing differentiated approaches based on proficiency level, structured teacher mediation for lower-level learners, and the complementary roles of AI and human instruction. Ethical considerations including data security and the risk of replacing rather than augmenting human instruction are addressed. The study concludes with practical recommendations for language educators seeking to integrate AI chatbots into their teaching practice.
A blessing or a curse?
Belmekki Asma1
Artificial intelligence is rapidly changing the way English as a Foreign Language is taught and learned in university classrooms. In Algeria, this change has created both opportunities and serious concerns. This study explores Algerian EFL university teachers' real views, experiences, and challenges regarding the use of AI in their classrooms. It particularly investigates whether AI is perceived as a blessing that supports teaching and learning, or as a curse that weakens students' autonomy, creativity, and effort. The study adopted a descriptive research design. Data were collected through a web-based questionnaire sent via email to EFL university teachers from different Algerian universities. Forty (40) teachers voluntarily responded to the questionnaire. The findings show that teachers view AI as an inevitable part of higher education, but they are concerned about students' overreliance on it and its negative effect on autonomy, creativity, effort, and critical thinking. The study concludes that AI is not harmful in itself, but its uncontrolled use may negatively affect EFL learning. Therefore, Algerian universities need clear guidelines to promote responsible, balanced, and critical AI use in EFL classrooms.
A focus on Algerian EFL students at Mostaganem university.
Belmokhtar Kheira1
Artificial Intelligence (AI) has inevitably become an essential criterion in English language learning, offering a myriad of tools that support students in developing their language skills. These tools, such as grammar checkers, translation systems, writing assistants, and vocabulary applications, are increasingly used by English as a Foreign Language (EFL) learners to enhance their academic performance. Although AI has become increasingly embedded in educational settings, limited research has examined how Algerian EFL students use these tools and perceive their effectiveness. This study investigates Algerian EFL students' attitudes towards the integration of AI in English language learning at Mostaganem university. Specifically, it explores the frequency of AI use, the language skills students seek to improve via AI tools, and their level of trust in AI-generated academic content. This study draws on a quantitative descriptive approach, using an online questionnaire administered to Licence and Master students in the English Department. The findings reveal a generally positive attitude toward the integration of AI in English language learning. AI tools are primarily utilized to support writing, grammar correction, translation, and vocabulary acquisition, reflecting their growing role in students' academic practices. Learners acknowledge the usefulness of AI in enhancing language performance and facilitating learning processes. Nevertheless, trust in AI-generated content is not unconditional; concerns regarding reliability, accuracy, and excessive dependence on technology continue to shape students' perceptions. The study contributes to the growing body of research on AI in EFL education by providing insights into the experiences of Algerian university students and offering recommendations for the effective and responsible use of AI technologies in higher education.
A study of student perceptions at the university of Oujda
Garti Rihab1; Jmad Sabrine2; Zizaoui Ikram3; Eddal Imane1
Moroccan higher education is rapidly transforming due to the development of technologies such as artificial intelligence, this technology which offers next opportunities to the field of language learning. However, the effect of this technology regarding oral communicative competency remains under-researched. The present study's aim is to investigate students' perceptions regarding AI's capacity to develop their communicative competencies. A quantitative design is employed, data is collected via an online questionnaire administered by 135 students. The Technology Acceptance Model (TAM) is used to assess adoption, and the CEFR guided the analysis of oral competency components.The findings revealed a positive perception of AI. Students mainly perceive AI as a collaborative tool to improve their oral communicative competency. The study concludes that AI is considered as a compagnon in the process of learning languages, not just as an optional tool.
A critical analysis of task vulnerability in the New Prospects textbook in the age of generative AI
Abdelali Berriche1
The arrival of Generative Artificial Intelligence (GenAI) poses a radical challenge to the pedagogical design of traditional English as a Foreign Language (EFL) textbook. This paper argues that GenAI induces a "Calculator Moment" in EFL, collapsing the necessary "productive friction" for language acquisition. The results of the study suggest that traditional EFL textbook tasks, especially those that focus on lower-order thinking skills, are highly susceptible to AI replacement. A detailed checklist was used to analyse all the six units of the Algerian Third Year Secondary textbook, New Prospects. This checklist is based on an extended theoretical model of Krashen's Input Hypothesis (i+1), Bloom's Revised Taxonomy and Vygotsky's Zone of Proximal Development. Results indicate questions aimed at Remembering, Understanding and Applying are highly susceptible to AI, reducing the i+1 gap to an unproductive i+0. This analysis provides empirical support for the need for a curriculum migration towards higher order cognitive tasks that require human agency, critical evaluation and contextualized communication, and, in so doing, makes the teacher into a designer of necessary productive friction.
A comparative study of traditional and technology-assisted learning preferences among Indian school students
Naveen Tokas1; Dhruv Kumar2; Aakriti Sahay2; Fahran Ali Khan2
In today's era, Artificial Intelligence and digital technologies are reshaping learning preferences in language education, particularly among Generation Z students. This study explores the comparative perspectives of Indian school students regarding technology-assisted foreign language learning and traditional classroom-based learning in the contemporary educational environment. The research focuses on students' learning preferences, engagement, and perceptions towards foreign language learning in schools.The study was conducted by Grade VIII and IX students, student-led research initiative, under the supervision of a guide, using an online survey questionnaire as a research tool. The survey was distributed to students from different regions of India. More than 2,000 responses were collected and analyzed using percentage analysis, frequency distribution, and graphical representation. The findings reveal that although Gen Z learners are growing up in the age of artificial intelligence, traditional methods of foreign language learning continue to have a significant impact on their learning experiences by improving understanding, interaction, and learner engagement. The responses further indicate a strong interest in learning foreign languages in India, such as French, Italian, and German, despite ongoing policy changes regarding foreign language education in Indian schools. The study highlights the importance of balancing traditional pedagogical practices with technology-assisted learning approaches in foreign language education. It also demonstrates the broader transformation of foreign language learning within AI-supported educational environments. The research contributes to current discussions on Gen Z learners, educational technology, and the evolving future of language education among school students. IndexTerms -Gen Z, Foreign Language Learning, Artificial Intelligence, Technology-Assisted Learning, Digital Education.
Purposes, benefits, and challenges
Ouadjeri Akila1
This study investigates EFL university students' use and perceptions of Generative Artificial Intelligence (GenAI) tools in English language learning contexts, focusing on their purposes of use, perceived benefits, and associated challenges. The study was conducted among students at the University Center of Barika, Algeria, using a quantitative descriptive research design. Data were collected through an adapted questionnaire based on the study of Almassaad, Alajlan, and Alebaikan (2024), which explored students' perceptions of generative AI technologies in higher education. The questionnaire examines students' awareness and familiarity with GenAI tools, the frequency of their use in English language learning, and the reasons behind either using or avoiding these technologies. It also identifies commonly used GenAI applications, including ChatGPT, Gemini, Copilot, and QuillBot. Furthermore, the study explores educational purposes for using GenAI tools, such as translation, concept clarification, literature search, summarization, idea generation, proofreading, assignment completion, project work, and multimedia presentation creation. The study also investigates students' perceptions of the benefits of GenAI tools, including ease of access, time efficiency, instant feedback, improved language proficiency, enhanced academic performance, increased learning engagement, and learner confidence. In addition, it addresses challenges associated with GenAI use, such as internet dependence, subscription costs, privacy and data security risks, unreliable information, reduced learning autonomy, decreased human interaction, plagiarism, cheating, and potential negative effects on learning. The study aims to provide an understanding of how GenAI tools are transforming English language learning experiences in higher education while highlighting pedagogical opportunities and ethical concerns related to their integration in EFL contexts.
ChatGPT between educational support and critical dependence in higher education
Wafia Tihal1
Fast development of artificial intelligence (AI) has revolutionized higher education. ChatGPT is among a number of AI technologies that are currently being extensively used for offering explanations, summaries, and aiding in writing (Kasneci et al., 2023; Lo, 2023). This study is motivated by an observed incident during a university graduation ceremony in Algeria, where a student wrote "ChatGPT" on her graduation cap. This symbolic gesture was interpreted differently by instructors. Some teachers viewed it as an implicit provocation or even an offence, suggesting that students were attributing their academic work, projects, assignments, and written tasks to ChatGPT rather than their own intellectual effort. Others interpreted it as a sign of the growing normalization of AI tools in students' academic practices and the increasing visibility of AI-assisted learning. The study adopts a qualitative exploratory approach based on interpretive analysis of teachers and students' reported academic practices regarding the use of ChatGPT in assignments and coursework. The results show a trend toward an increasing issue concerning academic integrity because the majority of students use ChatGPT to write papers but demonstrate little evidence of any critical engagement. Although the use of AI helps students understand things more easily, a clear trend is observable in terms of overreliance on the system's output. The study concludes that generative AI use ranges from academic support to dependency. It recommends clear institutional guidelines, integration of AI literacy, and assessment practices that promote critical thinking, process-based evaluation, and academic integrity.
A pedagogical framework for intentional prompt engineering in EFL material design
Krechich Fadoua1; Hairech Faiza1
The rapid integration of Generative Artificial Intelligence (GenAI) has transformed instructional design in English as a Foreign Language (EFL) education. While Large Language Models (LLMs) offer unprecedented efficiency in generating reading texts, vocabulary lists, and lesson plans, their mainstream educational usage often relies on superficial, ad-hoc prompting. This erratic approach frequently yields outputs that lack pedagogical scaffolding, cultural relevance, or alignment with established language acquisition frameworks. Consequently, there is an urgent need to shift the paradigm from passive, automated text generation to structured, intentional human-AI communication in curriculum development. This conceptual paper introduces a robust, teacher-facing framework designed to guide EFL educators in communicating systematically with GenAI tools. By treating prompt engineering as a form of specialized, pedagogical dialogue, the framework aims to ensure that AI-generated instructional materials strictly adhere to specific linguistic levels, task-based objectives, and communicative language teaching (CLT) principles. Utilizing a design-based conceptual approach, this study outlines a structured system-prompting matrix. The framework operates on four core linguistic constraints: target pedagogical framework alignment, lexicogrammar profiling, localized contextualization, and interactive scaffolding. To evaluate the framework's viability, a qualitative walkthrough is presented, illustrating how an instructor iteratively prompts an LLM to design a hyper-personalized, context-specific reading unit and an accompanying vocabulary assessment for advanced EFL learners. The generated artifacts are evaluated against traditional instructional design rubrics to determine their pedagogical soundness. The paper demonstrates that when educators engage in structured, constraint-driven dialogue with GenAI, the resulting materials exhibit higher instructional alignment and fewer algorithmic errors or generalizations. Ultimately, this framework contributes to the discourse on digital language policy by repositioning the teacher not as a bystander to automated automation, but as an active, agentic mediator of educational technology.
Nourelhouda Latreche1; Houda Bouhadjar1
The demanding nature of literary studies often poses challenges for teachers in literature classrooms. A vast number of literary texts are lengthy, complex, and require deep interpretation; thus, they require learners to engage deeply with the material in order to achieve the objectives of literature instruction and develop literary competence, all within limited classroom time. In this context, microlearning can offer focused and manageable learning activities that support engagement with literary texts while preserving their depth and nature. This conceptual paper explores the potential of AI-powered microlearning as an instructional approach in literature classes. Specifically, the paper examines how artificial intelligence can assist teachers in designing microlearning activities for literary themes, characterization, literary devices, and textual interpretation, while also considering the pedagogical opportunities and challenges associated with such integration.It aims to demonstrate how artificial intelligence can support the design of flexible and personalized microlearning activities that break literary texts into manageable learning units, such as character analyses, theme exploration tasks, and guided reflections, thereby assisting teachers in achieving the instructional objectives of literature classes.
A framework for teaching grammatical cohesion in text linguistics using Qwen Chat
Imane Bahri1
Since its emergence, artificial intelligence (AI) has generated considerable interest regarding its integration into language education. However, existing research has largely focused on how students utilize AI to assist with their own writing, paying limited attention to how language educators can actively incorporate AI into the explicit teaching of complex textual features. This paper proposes an AI-assisted pedagogical framework for teaching grammatical cohesion in text linguistics courses. The theoretical framework draws on Vygotsky's sociocultural theory of dynamic scaffolding and Halliday and Hasan's framework of grammatical cohesion (reference, substitution, ellipsis, and conjunction). Within this framework, large language models such as Qwen Chat are positioned not as substitutes for instructors expertise, but as flexible pedagogical scaffolds. Specifically, the framework demonstrates how instructors can employ AI to: (1) generate contrastive linguistic data (minimal pairs highlighting cohesive shifts), (2) analyse cohesive errors in learner texts, and (3) facilitate critical AI literacy by having students evaluate the semantic coherence of AI-generated cohesive ties. This framework is evaluated through a qualitative action research study involving 25 Master's 1 linguistics students in the English Department at Blida 2 University in Algeria, utilizing structured Qwen Chat prompts over a focused 3-week instructional module.] Preliminary findings indicate that when AI is strategically mediated by the instructor, it significantly enhances learners' meta-linguistic awareness of cohesive devices while preserving the teacher's fundamental role in guiding critical discourse analysis. This paper concludes with practical guidelines for linguistics educators seeking to integrate generative AI into advanced discourse analysis instruction.
A quasi-experimental study on the impact of AI on quiet EFL learners' classroom participation.
Matallah Taous1; Ouis Hanen2
In traditional classrooms, many shy students struggle to engage because of their fear of making mistakes. However, artificial intelligence (AI) can empower quiet EFL learners to engage in classroom instructions. The researcher highlights the importance of AI voice assistant tools in creating an enthusiastic approach. Accordingly, the researcher scrutinizes the use of AI chatbots and voice assistants to create a low-pressure environment where learners embrace their confidence and power to encounter their fears and feel at ease in sharing their ideas. Using a one-group pretest-posttest design aligned with an intervention, a group of EFL students with low levels of classroom participation was identified based on prior classroom performance. This study aims to determine whether interaction with AI tools may enhance learners' willingness to engage in classroom instructions. In the same vein, the findings are expected to provide insights into the impact of AI on creating an entertaining environment that boosts less active learners to indulge more confidently in language learning with a big smile and great desire. Accordingly, many future recommendations are proposed. AI tools should be embraced in EFL classrooms, especially in oral sessions. Besides, giving attention to shy low-confident students will enhance the teaching outcomes at the end of the educational course. In addition, an AI classroom should be linked with a traditional classroom, highlighting the crucial role of the teacher's guidance to ensure a balanced communicative development.
Adapting EFL writing pedagogy for the AI-mediated composition process
Radia Bouguebs1; Nadira Bouguebs2
The post-2022 paradigm shift in educational technology has transitioned Artificial Intelligence from a basic grammar-checking tool into an active partner in "generative authorship." This evolution challenges traditional notions of student agency, assessment integrity, and text ownership in English as a Foreign Language education. This paper presents an evidence-based pedagogical framework that merges deep cognitive learning theories with empirical qualitative data to address this disruption. Theoretically, the study adapts Vygotsky's Zone of Proximal Development, Cognitive Load Theory, and Self-Determination Theory to re-evaluate the human-AI collaborative writing loop. It frames the central tension as a choice between AI acting as a cognitive scaffold that aids developmental growth or a cognitive bypass that masks underlying learning gaps with sterile, algorithmic text perfection. Empirically, the paper draws on a Reflexive Thematic Analysis exploring university teachers' confidence, anxieties, and trust regarding the assessment of AI-mediated writing. The qualitative findings reveal significant fractures in traditional grading criteria, driven by an inability to distinguish developmental, "approximate language" from automated generation, which severely undermines teacher assessment confidence. To bridge this divide, this study proposes an Adaptive Writing Ecosystem that moves beyond screen-level prompt engineering to prioritize the internal cognitive and affective states of the writer. The framework advocates for a structural shift from product- to process-based rubrics, institutional guidelines separating brainstorming from drafting phases, and the integration of oral verification protocols. Ultimately, the study concludes that while technology personalizes the text, only humanized pedagogy can personalize the meaning-making process.
The role of voice-based AI in promoting speaking self-efficacy among EFL students
Fatma Zohra Boukehili1
The implications of artificial intelligence (AI) in education have extended beyond the linguistic domain to encompass the psychological dimensions of both learners and teachers. Nowadays, AI tools, particularly chatbots, are no longer limited to linguistic capabilities; rather, they show signs of emotional intelligence (EI). In the EFL context, while AI has shown real promise in supporting different areas of learning, there remains a noticeable lack of research examining its impact on learners' emotional state, especially in relation to spoken language development. Therefore, the present study seeks to investigate the role of an emotionally intelligent voice-based chatbot (Pi) in enhancing EFL students' speaking self-efficacy. The study mainly focused on three dimensions of self-efficacy: performance, self-regulatory and linguistic self-efficacy. Thirteen Algerian EFL students enrolled at the University of Souk-Ahras participated in the study through convenience volunteer sampling. The study employed a quantitative one-group pre-test-post-test pre-experimental design where a modified version of the Self-Efficacy for English Speaking Scale (SEESS) was administered as both a pre-test and a post-test to examine changes in students' speaking self-efficacy before and after the intervention. The experiment ran over seven sessions, each lasting about three hours. During each session, students engaged in spoken conversations with Pi on various topics and frequently asked it for feedback on their speaking. The findings revealed statistically significant improvements across all dimensions of self-efficacy, with the greatest gains observed in performance self-efficacy, followed by self-regulatory efficacy and linguistic self-efficacy, indicating that the AI tool enhanced students' confidence in performing speaking tasks, managing their own learning, and using the language effectively.
A theoretical examination of learner self-efficacy in academic writing
Khaoula Atmani1; Mehdi Kaddouri2; Abdellatif Belkadi3; Amal Zerouali2
The growing integration of AI-powered writing tools in higher education has introduced new and underexplored challenges for academic writing development. While these tools are considered as valuable feedback resources capable of enhancing students' writing quality, this presentation argues that their systematic tendency to suggest modifications to already correct and academically appropriate writing constitutes a form of overcorrection with serious pedagogical consequences. Unlike traditional corrective feedback, which targets genuine errors, AI tools such as ChatGPT and Gemini consistently propose alternative vocabulary, restructured sentences, and stylistic changes regardless of whether the original writing meets academic standards. This pattern of unnecessary intervention risks sending students a misleading message — that their language choices are perpetually insufficient. Drawing on Bandura's (1997) self-efficacy theory, Truscott's (1996) critique of written error correction, Hyland's (2002) academic voice framework, and Holec's (1981) learner autonomy theory, this presentation theorizes that persistent AI suggestion creates a false impression of inadequacy in students, which leads them to distrust their own linguistic judgments and undermine their confidence as academic writers. Constant AI intervention further erodes learner autonomy and silences the development of an authentic scholarly voice, carrying direct negative consequences for long-term academic writing performance. This presentation calls for raising critical awareness among teachers and students regarding the overcorrective nature of AI feedback tools, advocating for a balanced and reflective approach that encourages students to trust their own competence rather than developing an uncritical over-reliance on AI-generated suggestions.
A pilot study on AI-assisted speaking and foreign language anxiety
Firdevs Bozdemir1; Mehmet Aslan2
This pilot study, which adopts a mixed-methods research design, explores the instructional impact of Praktika, an artificial intelligence-supported speaking practice tool, on reducing foreign-language speaking anxiety among beginner-level high school students. Quantitative data were collected through an adapted version of the Foreign Language Speaking Anxiety Scale from Horwitz et al. (1986), administered as pre- and post-tests. Qualitative data were gathered through semi-structured interviews exploring learners' affective responses, confidence in speaking, and perceptions of Praktika. The participants were comprised of 12 ninth graders in a state high school in Türkiye. Throughout the four-week intervention process, the students participated in speaking activities with Praktika besides regular classroom activities. The findings gathered from the quantitative analysis pointed to a significant decrease in participants' foreign language speaking anxiety scores after the intervention. Qualitative results indicated that students reported enhanced speaking confidence, reduced fear of negative evaluation, and a tendency to accept errors as a component of the learning process. Participants attributed these improvements to Praktika's non-judgmental atmosphere, immediate feedback function, and opportunity to practice at their own pace. Although this research is constrained by a limited sample and a short-term intervention, it offers preliminary evidence that Praktika can function as a beneficial complementary tool to reduce spelling anxiety and foster speaking proficiency.
Teachers' attitudes and challenges in inclusive EFL classrooms
Khouane Abbassia Hanane1
In today's EFL classrooms, where learner heterogeneity is the defining characteristic, the implementation of one-size-fits-all instruction is no longer pedagogically viable. Designing interactive lessons that meet the multifarious needs of learners and their learning paces, in what is known as differentiated instruction, has been one of the essential yet daunting tasks for teachers, not to mention providing timely, tailored feedback. However, with the swift advancements of artificial intelligence in education, the implementation complexity might be minimized. Effectively integrating AI may help educators mitigate several challenges while offering personalized learning experiences that, in turn, may foster interaction patterns. This study is, therefore, an attempt to examine the major challenges that teachers encounter in inclusive classrooms. Moreover, it seeks to highlight their attitudes towards using AI as a co-assistant to alleviate their workloads. Furthermore, it aims to explore whether teachers who are already using AI in planning their lessons find any improvement in their teaching methods and in their classroom interaction. To reach this end, a mixed-methods case study will be employed, using an online questionnaire administered to EFL teachers from different Algerian universities, alongside classroom observations. The triangulation of the results is expected to shed light on the myriad difficulties that teachers face in adapting their instructional practices to their students' needs. The findings may also reveal that educators hold positive attitudes towards using artificial intelligence in reducing planning burdens while increasing interactive opportunities.
A systematic review of applications, outcomes, and challenges in higher education
Mohd Shahbaz Khan1; Sameer Babu M1
The rapid adoption of Artificial Intelligence (AI) into higher education is transforming pedagogical methodologies, especially in language instruction. English for Specific Purposes (ESP), a needs-oriented methodology centred on discipline-specific communication, is particularly shaped by these advancements. despite increasing interest, a systematic synthesis of empirical evidence regarding AI applications in ESP contexts is still scarce. This study offers a systematic review of research published from 2020 to 2026, adhering to the PRISMA framework. The review examines three primary inquiries: (1) What categories of AI tools are employed in English for Specific Purposes (ESP) within higher education? What learning experiences and outcomes have been documented? What advantages and obstacles arise from their utilization? A systematic search of the Scopus database identified 15 peer-reviewed studies that fulfilled the inclusion criteria. Data relevant to AI tools, research methodologies, educational outcomes, and implementation obstacles were systematically extracted and synthesized. The results demonstrate a rapidly changing AI environment, characterized by a significant increase in publications from 2024 to 2026, indicative of the ascendance of generative AI technologies. The findings indicate that the AI ecosystem in ESP has evolved from conventional corrective tools to generative AI applications like ChatGPT, in addition to specialized tools that facilitate research and oral communication. Secondly, learning outcomes are significantly influenced by pedagogical design: a structured and guided application of AI correlates with enhancements in writing proficiency, critical thinking, and learner autonomy, while unguided use may result in overreliance, diminished skill development, and issues related to academic integrity. Third, AI exerts a dual influence, providing personalized feedback, enhancing motivation, and alleviating language anxiety, while simultaneously introducing challenges concerning accuracy, disciplinary relevance, teacher-student perception discrepancies, and socio-emotional engagement. The review asserts that the efficacy of AI in ESP is contingent not solely on technology but also on its pedagogical integration.
Hafsa Abbou1; Ghobrini Rafik El Amine2; Zouar Houda2
This pedagogical action research explores the experiences of learning among master students with a particular focus on the integration of AI in oral expression module. While there is a growing awareness of the role of AI in enhancing language skills among English language university students, little is known about its use within the Algerian context to enhance students speaking skills particularly when English is becoming a primordial language in Algeria. This study was guided by sociocultural theory of Vygotsky as well as Gamification and motivation theory to explore the extent to which the learning experiences within oral expression module is mediated and supported through AI-generated games. To this end, observation and questionnaires were collected from 30 master students. The analysis of the data reported an increased students' engagement, peer collaboration and a greater willingness to participate in speaking activities. These findings suggest that integrating AI-driven gamified tools can create a more dynamic and supportive environment for developing oral proficiency in English. This study holds relevance for higher education contexts seeking to ethically integrate AI into Oral Expression classrooms, particularly in settings undergoing a linguistic shift where English is emerging as a primary language of instruction and communication.
Rasha Jadou1
As artificial intelligence (AI) becomes more central to global communication, protecting endangered and marginalized languages is becoming increasingly important. This paper argues that AI should not be used only for translation. Instead, it should serve as a tool that helps languages survive, grow, and stay active in the digital world. This means treating language technology as a long‑term investment, not just an optional technical feature. Today, more than 6,000 languages are missing from modern AI systems. This problem is not only caused by a lack of data. It also results from limited funding, scattered efforts, and the absence of strong, coordinated plans to support languages with fewer digital resources. To address this imbalance, we need AI strategies that reduce the gap between well‑supported languages and those that are digitally disadvantaged. This includes creating fairer algorithms and ensuring equal access to digital tools. A key part of this approach is building technology that centers on the communities who speak these languages. When local speakers lead the development of digital tools, the results are more culturally accurate, meaningful, and useful. Community‑driven design ensures that AI systems respect cultural knowledge and adapt to local needs. The paper also highlights the importance of collaboration. Protecting linguistic diversity requires partnerships between communities, researchers, cultural organizations, and international institutions. Language preservation is not only a technical task—it is a long‑term effort to protect cultural heritage. Through these combined efforts, AI can help reduce digital inequality and give endangered languages a stronger presence in the rapidly evolving digital world.
Opportunities, challenges, and ethical frontiers
Doaa Ahmad Hawa1
Linking Human Language and Artificial Intelligence: Opportunities, Chal Recent advances in artificial intelligence (AI) have significantly altered how machines understand and generate human language. Large language models have made it possible for natural language processing (NLP) systems to perform complex linguistic tasks like sentiment analysis, translation, summarization, and dialogue generation with near-human proficiency. This paper explores the dynamic relationship between language and Al, focusing on how computational models learn language, interpret meaning, and deal with contextual variation using large-scale datasets.The study looks at important technological developments that have spurred progress in this field, particularly transformer-based architectures and deep learning techniques. It also addresses enduring problems like training data bias, cultural and contextual knowledge gaps, and concerns about accountability and transparency in automated systems. The study also looks at how Al-driven language technologies may affect society more broadly in areas like digital communication, healthcare, and education.By critically analyzing both the advantages and disadvantages of existing Al language systems, this study advances our understanding of how these systems impact human interaction and knowledge production. In addition to highlighting future research directions, the paper's conclusion emphasizes the significance of developing inclusive, understandable, and morally sound Al systems that are consistent with human values.
Ashwaq Ahmad Hammash1
Artificial intelligence (AI) is increasingly influencing English language learning in higher education by providing innovative tools that support skill development and personalized learning. This paper examines the role of AI applications such as ChatGPT and language correction systems in improving learners' English proficiency, particularly in writing, vocabulary, and communication skills. The study highlights that AI enhances self-directed learning by offering instant feedback and flexible learning opportunities outside the classroom. It also increases student engagement and confidence in using the English language in academic contexts. In addition, the paper includes a brief international comparison of AI integration in English language education. It suggests that some educational systems have adopted AI more systematically and effectively, while others are still in the early stages of implementation despite growing interest. The findings indicate that AI can significantly improve language learning outcomes when used appropriately, but it should be balanced with traditional teaching methods to ensure critical thinking and academic integrity.
Pedagogical evidence, classroom realities, and human limits
Swati Arora1
This paper examines the integration of Artificial Intelligence tools into English Language Teaching (ELT) in Japanese university classrooms, focusing on what is pedagogically effective, what falls short in practice, and where human teaching remains irreplaceable. Drawing on classroom observation and educators, this study critically evaluates AI-assisted language learning tools, including writing assistants, automated feedback systems, and conversational AI, within the specific sociolinguistic and cultural context of Japan. Japan presents a distinctive case: a high-technology society with deeply rooted traditional pedagogical norms, where learner anxiety, cultural expectations around error and performance, and institutional conservatism shape how AI tools are received and used. The paper identifies pedagogical gains in areas such as grammar feedback, vocabulary reinforcement, and learner autonomy, while also documenting consistent limitations: AI tools frequently fail to address pragmatic competence, intercultural communication, and the affective dimensions of language learning that skilled teachers navigate intuitively. Crucially, this study argues that the boundary between productive AI assistance and over-reliance is not merely technical but ethical and pedagogical. Teachers must actively mediate AI use to preserve student agency, critical thinking, and authentic communicative development. The paper concludes with evidence-informed recommendations for ELT practitioners and institutional policymakers on how to integrate AI responsibly, sustainably, and in ways that keep human judgment at the center of language education.
Navigating the pedagogical friction of EMI in Algerian medical education
Lallaouna Mounir1; Ounis Salim2
While the recent adoption of English-Medium Instruction (EMI) in Algerian higher education aims for globalization, it has sparked significant pedagogical friction. This study moves beyond traditional perspectives to investigate "Shadow Didactics"- the survival strategies students use to navigate this shift. It posits that EMI is mediated through peer-led networks and AI tools, forming an informal pedagogy that bridges the gap between students' linguistic reality and policy demands. Employing a convergent mixed-methods case study at Abbas Laghrour University (2024–2025), the research focused on first- and second-year students in the faculty of medicine. The research used a questionnaire which measured self-reported proficiency, peer mediation frequency, Telegram and WhatsApp usage, and attitudes toward translanguaging. Additionally, an interview was conducted with the students to deepen the investigation.Preliminary findings reveal that academic success in this transitional phase results from a hybridized mixture of informal methods rather than English-only instruction. Peers and AI serve as focal mediators; proficient classmates act as pillars connecting less proficient peers to contentknowledge. Success in Algerian EMI classrooms depends on belonging to interconnected networks that offer immediate, contextualized mediation—a reflection of social competence rather than linguistic deficit. The study concludes that the role of the peer mediator is salient for translating complex terminology. Consequently, policymakers should move away from rigid models and adopt a translanguaging pedagogy that integrates peer-mediated scaffolding. "Shadow Didactics" serves as a critical alert: the gap between formal policy and pedagogical reality remains significant.
The impact of artificial intelligence on literary terms acquisition in the English classroom
Boutkhil Guemide1
This study examines the impact of AI-powered tools and applications on the learning and retention of literary terminology among 1st year EFL students of Msila University. The research employs a mixed-methods approach, combining quantitative data on student performance and qualitative insights from teacher and student perspectives. English classrooms were divided into control and experimental groups, with the experimental group utilizing AI-assisted learning modules focused on literary terms. Pre- and post-assessments were conducted to measure the acquisition and long-term retention of key literary concepts. Results showed the AI group achieved significantly higher post-test scores (M=18.2 vs. M=15.6, p<0.01) and retention scores after three months (M=17.5 vs. M=14.8, p<0.01). Qualitative data revealed enhanced engagement and confidence in the AI group. These findings suggest AI can enhance literary education by improving knowledge acquisition, engagement, and retention. In addition, the findings of the study suggest that the integration of AI in the learning process significantly improved students' understanding and recall of literary terms compared to traditional instructional methods. AI-powered applications provided personalized feedback, adaptive practice, and immersive learning experiences that enhanced engagement and knowledge retention. Moreover, the qualitative data revealed that both teachers and students perceived the AI-augmented approach as more effective, efficient, and enjoyable than conventional literary instruction. The study's implications highlight the potential of AI to transform the teaching and learning of literary analysis in EFL classrooms. By leveraging the strengths of AI, educators can foster deeper comprehension, increase student motivation, and better prepare learners for the demands of literary study.
Enhancing language learning through adaptive technologies
Kherroubi Hassiba1
The rapid development of Artificial Intelligence (AI) has transformed educational practices and created innovative opportunities for multilingual learning environments. This paper explores the role of AI-driven technologies in enhancing language acquisition, improving learner engagement, and supporting personalized education. Through adaptive learning systems, intelligent tutoring platforms, speech recognition tools, and automated feedback mechanisms, AI enables learners from diverse linguistic and cultural backgrounds to access more inclusive and effective educational experiences. The study examines how AI contributes to developing communicative competence, vocabulary acquisition, pronunciation accuracy, and intercultural understanding among students in multilingual classrooms. It also discusses the challenges associated with the integration of AI in education, including ethical concerns, data privacy, algorithmic bias, and the potential reduction of human interaction in language learning processes. By analyzing recent educational applications and digital learning practices, this paper argues that AI can serve as a valuable pedagogical tool when implemented responsibly and collaboratively with educators. The research highlights the importance of balancing technological innovation with human-centered teaching approaches to ensure equitable and meaningful learning experiences. Ultimately, the paper demonstrates that AI has the potential to reshape multilingual education by fostering accessibility, personalization, and intercultural communication in the digital age.
Learners multimodal apprenticeship in vlog-based oral communication
Nurul Nazira Binti Hamzah1; Esther Binti Jawing2
Video-based multimodal production and generative artificial intelligence (GenAI) tools are increasingly reshaping how learners plan, perform and refine multimodal texts in language education. While vlogging and GenAI chatbots have individually gained traction in ESL research, limited empirical work has examined how learners experience their combined use within vlog-based oral communication tasks. Addressing this gap, the present study explored Malaysian ESL undergraduates' written reflections following a semester-long vlog-based English oral communication project where generative AI (GenAI) were introduced as an optional support tool. Drawing on reflective thematic analysis of 34 reflections, the findings illustrate how learners engaged with multimodal apprenticeship encompassing idea generation, script structure, refining language and supporting production decisions, while selectively adapting AI output to maintain personal voice and authenticity. Interpreted through the Self-Determination Theory (SDT) and Zone of Proximal Development (ZPD), the analysis shows how GenAI chatbots functioned as a form of cognitive scaffolds and co-creators that supported autonomy, competence and relatedness; and confidence without replacing human modelling. Peer collaboration, online examplars and teacher guidance remained central sources to the learning process. At the same time, learners demonstrated emerging critical AI literacy by recognising limitations related to accuracy, reliability, time and technical constraints. Overall, the findings challenge deficit narratives that portray GenAI as undermining learner agency, instead of highlighting its role as a supportive tool within a guided multimodal apprenticeship.
A case study of academic staff at Belhadj Bouchaib University
Khalladi Meriem1
Since artificial intelligence continues to reshape higher education, international academic mobility no longer requires linguistic competence alone. Academic staff, including teachers, researchers, and students involved in mobility programmes, are expected to develop intercultural, digital, and social skills that enable them to communicate and adapt effectively across diverse academic environments. Similarly, English language training curricula should move beyond traditional language instruction and respond to the communicative, cultural, and technological realities of contemporary higher education. Nevertheless, many curricula implemented in home universities still focus primarily on general English proficiency, with limited attention to AI-mediated communication, intercultural preparedness, and the social experiences encountered in host universities. This study explores the extent to which AI-enhanced English language curricula can support academic and intercultural integration during international mobility experiences. It examines the relationship among curriculum design, digital transformation, and the communicative challenges participants face as they transition from home to host university settings. Particular attention is paid to the roles of artificial intelligence tools, technology-enhanced learning environments, and innovative curriculum practices in developing intercultural awareness, academic communication skills, and social adaptability in multilingual, digitally mediated educational contexts. The study uses an online questionnaire that aims to investigate the perceptions and experiences of university mobility participants at Belhadj Bouchaib University in Ain Temouchent regarding the effectiveness of existing English language training programmes. The findings are expected to reveal the importance of developing more context-sensitive, AI-oriented curricula that integrate authentic mobility experiences, intercultural communication practices, and digital competencies aligned with the evolving demands of global higher education.
Students', parents', and teachers' perspectives on AI-supported education and language learning
Md. Saddam Hossain1
The use of generative artificial intelligence for learning practices is increasing day by day. Now, Bangladeshi students, parents, and teachers are using ChatGPT, and AI tool, for academic support, language learning, writing assistance, problem solving, and knowledge verification. This study explores how ChatGPT is integrated into Bangladeshi educational practices, with particular attention to its role in AI-supported education, multilingual learning, and subscription-based access. Using an exploratory mixed-methods approach, structured questionnaires with Likert-scale items, multiple-choice questions, and open-ended responses were administered to three key user groups: students, who use ChatGPT for different purposes, including writing essays and applications, finding answers of subject-based questions, practicing English grammar, pronunciation and vocabulary, and solving critical mathematical problem, etc.; parents, who use it to respond to children's questions at home; and teachers, who employ it for preparing lesson, , making hand notes, providing explanation, feedback, and classroom support. The study examines perceived opportunities, challenges, ethical concerns, and the impact of free versus paid access on learning equity. Findings show that ChatGPT can play a great role for learning languages. Besides, it can act as a digital learning companion across home and classroom contexts. However, its use raises concern about overdependence, accuracy, academic integrity, digital literacy, and unequal access. By situating ChatGPT within the Bangladeshi learning ecosystem, this study contributes to current debates on responsible AI use, equitable access, multilingual education, and pedagogical transformation in the Global South.
A Human-Centred approach to English language teaching
Khawla Mohammed Al Hamoud1
Artificial intelligence is becoming an important topic in English Language Teaching, especially as students and teachers increasingly use tools such as ChatGPT, translation applications, writing assistants, and speech recognition systems. My proposed paper, "From Artificial Intelligence to Augmented Teaching: A Human-Centered Approach to English Language Teaching," explores how AI can support English language teaching without replacing the essential role of the teacher. The paper argues that AI should be understood not as a threat to teachers but as a tool that can enhance and extend the teaching-learning process. The study focuses on augmented teaching, in which AI helps learners improve their writing, pronunciation, vocabulary, speaking skills, and communicative competence through individualized practice and immediate feedback. It also considers how AI can assist teachers in preparing materials, designing classroom activities, and responding to different learner needs more effectively. However, the paper also addresses the challenges of AI use, including plagiarism, learner overdependence, inaccurate information, privacy concerns, and the possible weakening of students' critical thinking. The main argument of this paper is that successful AI integration in English Language Teaching depends on teacher guidance, ethical awareness, clear classroom policies, and critical digital literacy. The English teacher remains central in motivating learners, managing classroom interaction, interpreting students' needs, and developing cultural and communicative competence. Therefore, this paper concludes that the future of English Language Teaching should be teacher-guided and AI-supported, preserving the human values of education while benefiting from the possibilities of artificial intelligence.
A TAM-extended analysis of foreign language anxiety, trust, and learner experience in the Tunisian context
Leila Najeh1
Artificial intelligence has been gradually integrated into foreign language learning settings, mainly with the use of generative AI tools like ChatGPT, automated feedback systems, and language learning applications. While these technologies offer new pedagogical opportunities, their influence on learners' emotional experiences, particularly foreign language anxiety, has not been adequately examined in the Tunisian context. This study examines the effect of artificial intelligence on foreign language anxiety of Tunisian EFL learners who study technical majors. In particular, it discusses the opportunities and challenges of AI-assisted language learning. The current study was conducted on 60 Tunisian students enrolled in technical majors . The questionnaire was developed based on the Technology Acceptance Model (TAM) extended with the trust and anxiety constructs to investigate learners' perception, attitude, and experiences with ChatGPT in English language learning. The study aims to find out if AI-based interaction helps to reduce anxiety by providing personalized feedback, flexible learning environments, and a safe space for communication, or if it generates more stress by creating dependence on technology, reducing human interaction, and facing challenges of precision and academic integrity. The expected findings indicate that artificial intelligence may have a positive impact on learners' confidence and participation by reducing the fear of negative evaluation and promoting autonomous learning. But it also anticipates a number of pedagogical and ethical problems associated with over-dependence on AI technologies. The study contributes to the ongoing debates on the role of AI in language learning by emphasizing the affective aspect of technology integration and highlighting the importance of responsible and balanced AI implementation in the Tunisian EFL educational context.
Opportunities, ethical challenges, and pedagogical implications
Ahmad Jabr1
Generative artificial intelligence (AI) is rapidly reshaping English Language Teaching (ELT) by offering new possibilities for personalized learning, automated feedback, material development, and communicative practice. Tools such as AI chatbots, text generators, and writing assistants can support learners in developing vocabulary, grammar, writing fluency, pronunciation awareness, and intercultural communication skills. For teachers, generative AI may reduce workload by assisting with lesson planning, assessment design, content adaptation, and differentiated instruction. However, its integration into ELT also raises significant ethical challenges, including academic dishonesty, data privacy, algorithmic bias, overdependence on technology, unequal access, and the authenticity of learner-produced work. These concerns require careful pedagogical planning rather than uncritical adoption. This paper examines the opportunities, ethical risks, and pedagogical implications of generative AI in English language education. It argues that generative AI should be used as a supportive tool rather than a replacement for teachers, with emphasis on critical digital literacy, transparent assessment practices, responsible use policies, and learner autonomy. The study highlights the need for teachers to develop AI literacy and redesign classroom practices to balance innovation with ethical responsibility. Ultimately, generative AI can enhance ELT when implemented thoughtfully, critically, and inclusively within clear pedagogical frameworks.
Preserving pedagogical values and teacher-student relationships via the Human-Centered writing triangle
Fatima Zohra Bekheira1; Kouider Makhlouf2
The rapid integration of AI-powered writing tools into educational settings is reshaping how EFL students write and communicate within the classroom. While AI is increasingly praised for enhancing efficiency and language support, excessive dependence on AI-mediated writing may gradually replace the human pedagogical qualities that make learning transformative. This paper suggests that overreliance on AI writing tools may lead to classroom silence, diminished student voice, reduced intellectual confidence, weakened peer interaction, and the loss of productive struggle for writing development. Growing dependence on AI-generated language and immediate responses may shift students' trust away from their own thinking, creativity, and linguistic identity toward artificial systems, and may disrupt teacher-student relationships and the social ecology of the writing classroom. If EFL students interact more with AI systems than with teachers or classmates throughout the writing process, core pedagogical elements such as dialogue, empathy, collaborative learning, feedback exchange, and emotional support may become progressively compromised, raising a critical question about whether future AI-mediated classrooms will educate human thinkers or learners shaped by machine-oriented cognition. Grounded in Humanistic Theory and Social Constructivism, the paper introduces a practical conceptual framework entitled "The Human-Centered Writing Pedagogy Triangle," which positions AI as a complementary educational aid rather than a replacement for human interaction and pedagogy, emphasizing a balanced relationship between teachers, students, and AI to rehumanize classroom relationships and preserve central pedagogical values within technology-mediated learning environments.
Diana Hornoiu1; Alina Buzatu1; Veronica-Cristina Nedelcu1
Situated at the intersection of contemporary theoretical debates on intelligent technologies and the humanistic imperative to rethink academic education, this study examines the blind spots of artificial intelligence - what AI does not see, or it sees only partially - in order to assess the conditions of its pedagogical legitimacy. We argue that new theoretical, critical, and pedagogical instruments are necessary, but that their development must be accompanied by a rigorous concern for what remains specific to human thought: hermeneutic subtlety, critical discernment, pragmatic sensitivity, and intercultural understanding. The first case study examines literary and fictional texts, showing that AI tends to neutralise ambiguity, reduce symbolic density, overlook aesthetic singularity, and transform interpretation into stereotypical paraphrases. The second case study focuses on pragmatics, where AI frequently fails to calculate implicature, grasp irony, humour, interpret silence, indirectness and culturally coded forms of interaction. The third case study addresses Romanian as a foreign language, where these limitations become particularly visible in contexts involving intercultural mediation, idiomatic expression, pragmatic politeness, aspectual nuance, register variation, and the symbolic value of culturally embedded texts. Reflecting together on these blind spots is crucial for the formation of younger generations and early-career researchers. Rather than being regarded merely as operational limitations of intelligent technologies, the blind spots of AI must be consciously identified, critically examined, and methodologically integrated into academic disciplines. Students need to distinguish algorithmic processing from human thought: how algorithms generate, classify, and predict information, and how human understanding interprets, contextualises, evaluates, questions, and assumes responsibility for meaning. In this sense, the blind spots of AI should become part of the very process through which students learn to read, interpret, assess, contextualise, and produce knowledge. Such an approach transforms AI literacy into a broader pedagogical competence: the capacity to understand its limits, to question its outputs, and to preserve the contextual, ethical, pragmatic, symbolic, aesthetic, and intercultural dimensions of human understanding.
Ninuk Lustyantie1; Yunilis Andika1; Anim Putranto1; Muhamad Arbian Fahriza Putra1; Falisha Dinar Rustandi1; Tengku Ratna Soraya2; Dian Damayanti3
This study aims to examine the contribution of Français Immersif A1, a Virtual Reality (VR) and Artificial Intelligence (AI) learning environment, in improving A1-level French listening comprehension skills in Indonesia. The focus is directed toward pedagogical innovation, immersive experiences, and support for equitable quality of foreign language learning aligned with SDGs Methodology: A mixed-methods sequential design was employed following the ADDIE instructional framework (Analysis, Design, Development, Implementation, Evaluation) at Universitas Negeri Jakarta and Universitas Negeri Medan with A1 French learners. Quantitative data from pre-post listening tests were analyzed using inferential statistics. Qualitative data from semi-structured interviews, classroom observations, and policy document analysis were analyzed thematically. Results: Quantitative findings revealed a statistically significant improvement in listening comprehension with a very large effect size. Qualitative thematic analysis yielded three major impacts aligned with SDGs: First, VR immersion provided authentic, contextualized visual environments that reduced comprehension anxiety and enhanced inferencing skills (SDG 4 – Quality Education). Second, AI-powered adaptive feedback with speed control and multimodal scaffolding functioned as a non-judgmental "patient tutor," reducing the affective filter for anxious learners (SDG 9 – Responsible Innovation). Third, peace education-infused dialogues modeled non-violent communication and intercultural respect, developing empathy and prosocial behaviors that transferred to daily interactions (SDG Target 4.7 – Peace Education). The system operationalized all three principles of Indonesia's national AI policy—safe, inclusive, and ethical. Significance: This R&D study demonstrates that VR-AI frontiers, grounded in ADDIE methodology, transform digital learning frictions into cognitive, affective, and intercultural gains, advancing SDGs through inclusive, innovative, and peaceful education.
Integrating AI tools into anglo-american literature teaching — a georgian–european comparative perspective
Tamari Iremadze1
The growing presence of artificial intelligence tools in educational settings raises urgent questions not only about pedagogy and assessment, but about the future of literary reading itself. This paper examines the practical and theoretical implications of integrating AI language tools into the teaching of Anglo-American literature across secondary and tertiary educational contexts in Georgia, drawing on empirical classroom experience and a dual perspective as both a secondary school English teacher and a university lecturer. The paper situates this inquiry within a comparative Georgian–European framework. Georgia occupies a distinctive position between post-Soviet educational legacies and active integration into European pedagogical networks — including eTwinning, within which the author has led internationally recognised projects. This cross-contextual vantage point illuminates how educators in European partner countries navigate AI adoption within curricula shaped by both local traditions and supranational frameworks, offering perspectives of broader relevance to the EU and Balkan educational space. The paper argues that AI tools, when approached critically and pedagogically, can deepen student engagement with Anglo-American literary texts while simultaneously fostering critical thinking and media literacy. It also addresses attendant risks: the flattening of interpretive nuance, overreliance on machine-generated readings, and the erosion of student hermeneutic agency. Grounded in Universal Design for Learning principles and inclusive pedagogy, the paper proposes a framework for responsible AI integration that keeps the human reader — and the text — at the centre. Implications for curriculum policy, teacher professional development, and assessment integrity are discussed.
Case of Algerian high school EFL teachers.
Romaissa Ouail1
The rise of AI-powered Large Language Models (LLMs) such as Chat GPT, Gemini, and Claude has opened novel pedagogical possibilities in English as a Foreign Language (EFL) Education. Indubitably, the use of LLMs in EFL lesson planning is increasing, yet the cultural appropriateness of AI-generated/enhanced lesson plans has received limited attention from both teachers and scholars. As such, a discussion is due about how teachers evaluate the extent to which their AI-generated lesson plans' content, examples, names, norms, values and practices align with Algerian high school students cultural background in order to mitigate the risk of confusion, offense or disengagement among students, prior to the actual lesson implementation. As a result, this mixed method study examines how 20 Algerian high school EFL teachers evaluate and mediate lesson plans designed or enhanced with the use of AI tools namely LLMs. Data were collected through document analysis of paired materials; the initial AI output and the teachers' final version used in class. Informed by document analysis results, semi-structured interviews with few EFL teachers will follow for further explanation of the matter. The present study concludes that teacher evaluation of lesson plans'cultural appropriateness is essential for customizing them to local contexts and eventually for enhancing the overall pedagogical practice. Moreover, it highlights implications for EFL teacher training on critical AI literacy as well as cultural adaptation.
Ouchene Nourelhouda1
Artificial intelligence is increasingly influencing English as a Foreign Language teaching, especially in vocabulary learning. This paper examines how AI-based tools can support vocabulary development in EFL classrooms by helping learners practice new words, receive instant feedback, and study vocabulary according to their individual needs. AI applications such as chatbots, vocabulary learning apps, automated feedback systems, and adaptive learning platforms can provide learners with repeated exposure to words in meaningful contexts. They can also help teachers design activities, track learners' progress, and identify vocabulary difficulties more easily. However, the use of AI in vocabulary teaching also raises some concerns, including overdependence on technology, unequal access to digital tools, and the need for teacher guidance. The paper argues that AI should not replace the teacher, but should be used as a supporting tool that makes vocabulary instruction more interactive, personalized, and engaging. It concludes that the effective integration of AI in EFL classrooms requires clear pedagogical aims, teacher supervision, and a balanced combination of traditional teaching methods and digital learning tools.
AI tools and assessment practices in Algerian EFL classrooms
Amira Sid1
The rise of AI writing tools has fundamentally changed the way university students write, raising urgent questions about authenticity, academic integrity, and the future of writing assessment in higher education. This shift is particularly felt in EFL contexts, where writing already represents one of the most challenging academic skills for language learners. This study examines how EFL teachers at Souk Ahras University perceive and respond to the growing influence of AI on student writing and its implications for assessment practices. Through semi-structured interviews with EFL instructors, the study explores teachers'observations of changes in student writing, the challenges AI-generated texts create for fair evaluation, and how teachers are adapting their assessment practices in response. Preliminary findings suggest that teachers are increasingly encountering written work that is linguistically polished yet lacking genuine student voice and original thinking. Teachers also report growing uncertainty about how to evaluate AI-assisted writing fairly and consistently, and many express a need for clearer institutional guidelines and professional development support. Furthermore, findings indicate that AI use is prompting teachers to reconsider what writing tasks they assign and how they design assessments that genuinely capture student thinking. The study argues that meaningful responses to AI in writing education require more than detection tools — they demand a fundamental rethinking of writing pedagogy and assessment in Algerian EFL higher education.
Evidence from an Algerian teacher training institution
Dalia Itak Torche1
The rapid advancement of Artificial Intelligence (AI) is reshaping educational practices and opening new avenues for language learning. In English as a Foreign Language (EFL) contexts, AI-powered tools such as ChatGPT are increasingly drawn upon to support idea generation, content organization, and writing development. As these technologies become more widely accessible, understanding how ready students are to integrate AI into their learning has grown into a pressing concern. This study investigates Algerian EFL students' readiness for AI-supported writing practices and examines their perceptions of AI as a resource for writing development. Using a quantitative descriptive design, an online questionnaire was administered to 32 first-year EFL students at ENS Assia Djebar of Constantine. The instrument explored students' perceptions of writing difficulties, their current use of AI in writing-related tasks, and their openness to AI-supported learning practices. The findings reveal that writing is widely regarded as a demanding skill, with idea generation and organization standing out as particular pain points. Students were found to hold positive attitudes toward AI and reported drawing on AI tools to support their writing. They perceived these tools as helpful for improving writing quality, developing ideas, and refining written content. Participants also expressed a clear willingness to incorporate AI-supported approaches into future writing tasks, pointing to a growing acceptance of AI integration in language learning settings. Taken together, the findings underscore the potential of AI-supported writing practices as an innovative, learner-centered response to writing challenges in EFL contexts, while offering practical pedagogical insights into the meaningful integration of AI in higher education.
Fostering critical awareness, AI literacy and digital citizenship among EFL students f English department, Mostaganem, Algeria
Malika Benyoub1
Digital media literacy remains a persistent challenge in Algerian EFL classrooms, particularly when it comes to navigating artificial intelligence (AI) and algorithm awareness. Despite high exposure to technology, many learners find it difficult to evaluate automated content critically and responsibly. This study aims to examine the cognitive and ethical barriers that affect students' academic research performance in English, with a particular focus on AI literacy and the development of twenty first-century skills (the 4Cs: Critical Thinking, Communication, Collaboration, and Creativity). To achieve this, a mixed-methods approach was employed, combining student questionnaires and classroom observation checklists. The data were collected from forty-two students (N=42) in the English Department at Abdelhamid Ibn Badis University in Mostaganem. The analysis reveals patterns of uncritical reliance on generative outputs, algorithmic dependency, and gaps in digital ethics awareness, as well as observed improvements in structured collaborative tasks. Rather than providing a single, universal explanation, the study explores the intricacies of critical literacy in real classroom settings and interactive workshops. The results confirm the hypotheses that AI and algorithmic awareness have significant effects on students' critical evaluation skills, which in turn affect their overall learning outcomes. In this research, the importance of explicit and context-specific digital pedagogy is highlighted and the foundation for improved instructional and curriculum design in Algerian settings of English language teaching is laid.
Opportunities, challenges and ethical implications
Agam Kumar Verma1; Mehak Jonjua2
- Artificial Intelligence (AI) has significantly affected communication, education, and learning methods in higher education institutions around the world. The use of technology that includes intelligent tutoring systems, chatbots, adaptive learning programs, and generative artificial intelligence has brought new possibilities into play. However, the institutions involved in higher education involve individuals representing different generations with varied digital literacies, technophobia, and communication styles. Therefore, this paper aims to examine the involvement of AI in human communication between generations in higher education institutions internationally and especially in the Global South region. The paper analyzes existing literature on AI-driven learning, digital communication, and intergenerational interaction using an interpretative and theoretical methodology. While special emphasis is placed on the experience gained in developing countries such as India, it places the analysis within the context of wider international discourse on the role that AI could play in universities. This research considers ways in which AI can be used to facilitate better communication, promote inclusiveness in learning environments, and foster cooperation between different participants of the education process. In addition, it raises ethical questions in terms of privacy, bias, digital divide, academic cheating, and dependence on technology. In light of the findings presented, artificial intelligence is able to serve as an efficient intermediary of education communication as long as there is digital literacy, readiness, and human-focused approaches to teaching and learning. The contribution of this paper to the global debate on responsible applications of AI in education lies in its emphasis on the relationship between technology and social responsibility.
A pedagogically mediated framework for developing communicative competence
Athmane Boufes1
The rapid proliferation of generative artificial intelligence has transformed the landscape of language education, creating new possibilities for multimodal meaning-making, learner engagement, and communicative practice. Nevertheless, current discussions frequently oscillate between technological enthusiasm and concerns regarding automation, often neglecting the indispensable pedagogical role of the teacher in orchestrating meaningful learning experiences. This paper proposes a pedagogically mediated framework that integrates human and artificial intelligence creative expressions within multimodal English Language Teaching (ELT). Drawing on multimodal pedagogy, sociocultural theory, and contemporary perspectives on human–AI collaboration, the framework conceptualizes artificial intelligence not as a substitute for human agency but as a complementary creative and cognitive resource operating under informed pedagogical mediation. Through the integration of AI-generated textual, visual, auditory, and audiovisual resources into communicative learning tasks, the proposed framework seeks to enhance learners' communicative competence while fostering creativity, critical thinking, interaction, and learner autonomy. Methodologically, the study adopts a qualitative-prevalent mixed-methods orientation and reports the iterative development of the framework through successive design and refinement phases. Particular attention is devoted to the balance between technological affordances, ethical responsibility, and educational objectives in order to ensure meaningful and responsible AI integration. By advancing a human-centered approach to multimodal language instruction, the study contributes to ongoing debates concerning the pedagogical implications of artificial intelligence and offers a theoretically grounded model for the design of communicative learning environments in the emerging era of human–AI co-creativity.
Opportunities, challenges, and ethical considerations
Mechraoui Nouha Manal1
The growing integration of artificial intelligence (AI) into educational settings is reshaping how students learn, interact with information, and develop academic skills. In English as a Foreign Language (EFL) higher education, AI-powered tools such as conversational agents, writing assistants, and automated feedback systems offer new opportunities for supporting self-regulated learning. These technologies can assist learners in setting goals, monitoring progress, managing learning strategies, and receiving immediate feedback. However, the increasing reliance on AI also raises important pedagogical and ethical concerns related to academic integrity, learner autonomy, data privacy, and critical thinking. This conceptual paper explores the relationship between artificial intelligence and self-regulated learning in EFL higher education. Drawing on recent literature from educational technology, language learning, and AI-assisted instruction, the paper examines how AI tools can support the key dimensions of self-regulation, including planning, monitoring, reflection, and self-assessment. It also discusses potential risks associated with excessive dependence on AI-generated content and highlights the need for responsible and ethical integration of these technologies into language education. The paper argues that AI should be viewed not as a replacement for human learning processes but as a supportive tool that can enhance learner autonomy when used appropriately. Finally, recommendations are proposed for educators, institutions, and researchers seeking to maximize the benefits of AI while preserving the fundamental goals of language learning and student development.
EFL students' use of AI tools for understanding academic discourse
Lasmeur Nihad1
The rise of artificial intelligence (AI) in higher education has transformed the way learners engage with academic discourse and interpret academic content. In EFL (English as a Foreign Language) contexts, students rely on AI-powered systems to facilitate academic discourse comprehension, including lectures and academic texts. This leads to AI functioning as a mediator that support learners' comprehension. This study explores the role of human-AI mediation in higher education by investigating EFL university students' use of AI tools for understanding academic discourse. Specifically, it examines students' attitudes, perceptions and experiences regarding AI effectiveness in facilitating academic discourse comprehension by simplifying academic language, explaining complicated concepts, and fostering independent learning. This research also seeks to identify the challenges students face regarding accuracy when using AI as a mediator in academic settings. This study adopted a descriptive qualitative approach, employing a structured questionnaire administered to EFL university students at ENS "Assia Djebar" of Constantine, Algeria. It explored the students' attitudes toward AI tools for understanding academic lectures, their experiences with these technologies, and the challenges encountered regarding precision. The findings suggested frequent use by students of AI as a mediator in facilitating discipline- specific terminology encountered in their academic texts, summarizing lectures, and clarifying complicated concepts in academic content. Moreover, the results displayed an overall positive attitude towards the use of AI tools for understanding academic discourse. However, the findings displayed a degree of critical awareness as students expressed concerns regarding the accuracy and precision of AI-generated responses. This work prompts ongoing discussion surrounding human–AI communication and mediation in educational environments with a focus on a dual perspective of both finding value in human-AI mediation and using it cautiously in academic contexts.
Omenia Meriem Djafri1; Nawal Mebitil2
In an increasingly globalised and digitally mediated hospitality sector, effective communication is a fundamental competence for frontline hospitality employees, as they are expected to communicate effectively with guests from diverse linguistic and cultural backgrounds, often using English as a lingua franca. Hence, recent advances of educational technologies, particularly gamification and interactive digital AI-powered learning platforms, are regarded as promising tools for enhancing frontline employees' communication skills through providing engaging and practice-oriented learning environments. This paper seeks to synthesize existing research on gamification, AI-enhanced interactive learning platforms, and technology-mediated communication training to highlight their contribution in enhancing communication skills among frontline hospitality professionals within English for Professional Purposes (EPP) training. The data are gathered through a critical review of the existing literature on gamification in language learning, AI-driven platforms in EPP training, and communication training in hospitality to explore how these technologies can supportprofessional communication. The review of previous research suggests that gamification and AI-powered interactive platforms have the potential to reshape communication training in hospitality contexts. Gamification elements such as missions, rewards, and progress systems appear to significantly enhance learners'-workers' motivation and persistence in online training contexts. Interactive AI-driven digital platforms provide opportunities for experiential learning through simulations and scenario-based activities that mirror real-world hospitality encounters. The integration of gamified mechanisms with interactive AI-powered platforms fosters learner autonomy and engagement, and contributes to more dynamic and context- relevant training. This study recommends incorporating gamification principles and interactive digital platforms into hospitality EPP training programmes to create more engaging and practice-oriented learning environments and to bridge the gap between theoretical language education and practical workplace proessional communication. This research contributes to ongoing discussions on AI-enhanced and technology-mediated EPP training and highlights implications for hospitality professionals, EPP practitioners, course designers, and other stakeholders involved in EPP training programmes.
A conceptual analysis based on Byram's ICC framework.
Chafiaa Chebbat1; Fouad Boulkroun2
Language teaching and learning has known a remarkable change with the integration of generative AI (GenAI) and large language models (LLMs). While much research has focused on developing linguistic proficiency through GenAI, less attention has been given to their role in enhancing intercultural communicative competence (ICC). Grounded in Byram's model of ICC, this conceptual paper examines the potential of GenAI tools in intercultural communicative language teaching, addressing a critical gap in research on the intercultural dimensions of AI-mediated language learning. Two research questions are put forward: (1) how can GenAI tools and LLMs enhance the learner's ICC as introduced by Byram's model? (2) What are the ethical and cultural issues that may arise from the integration of generative AI tools into intercultural language teaching? The study adopts a conceptual analytical design based on a systematic review of GenAI tools in language education. It uses a conceptual analysis to examine the incorporation of these tools in relation to Byram's five dimensions namely attitudes, knowledge, skills of interpreting and relating, skills of discovery and interaction, and critical cultural awareness. This paper also highlights both the opportunities and the issues associated with integrating GenAI and LLMs into intercultural communicative language teaching, including algorithmic bias, cultural stereotyping, hallucination, and cultural homogenization. The paper concludes by emphasizing the need for critical engagement with GenAI tools, and the design of intercultural tasks that support the cognitive processes essential for intercultural language learning. It conceptualizes GenAI as a complex tool that can support intercultural language learning, yet its potential depends mainly on careful integration to reduce cultural bias and ethical risks.
Walaa Shekho1
MAS University Dr. Walaa Shekho, Ph. D Abstract Artificial Intelligence (AI) has developed to be a transformative force in English for Specific Purposes (ESP) training, shifting the focus from passive understanding to active, appropriate and contextual discourse. This research examines the reception and efficacy of generative AI tools as a primary method for enhancing conversation skills within highly specialized professional and academic spheres. While old traditional ESP frameworks often struggle to provide scalable, authentic domain-specific dialogue, modern AI-driven conversational agents offer personalized, real-time interactive frameworks. following a mixed-method approach, this study explores how learners perceive and apply these virtual interlocutors to practice industry-specific terminology, simulate high-stakes workplace scenarios, and receive immediate feedback as a correction on conversational register and fluency. The findings suggest that AI tool reception is largely helpful, significantly increasing learner self-sufficiency, reducing foreign language anxiety, and enhancing contextual communication confidence. However, the data also highlights persistent challenges, particularly concerning algorithmic generalization, a lack of human-to-human cultural orientation, and the risk of over-dependence. Ultimately, this paper argues that while AI tools work as powerful mechanisms for linguistic reception and conversational skills, they are most effective when arranged as supplementary accelerators rather than complete replacements for human-guided expert interaction. The study ends up with methodological recommendations for instructors seeking to improve human-AI collaboration within contemporary ESP course books.
Madina Kushukova1
Artificial Intelligence (AI) has significantly influenced secondary education. By incorporating AI, secondary school teachers of foreign languages have opportunities to raise the standards of instruction. In English language teaching, several longstanding challenges persist, particularly in traditional, teacher-centered approaches that often lack interactivity and real-time feedback, especially in rural schools. The advent of mobile learning platforms and AI-driven technologies offer a promising solution to these challenges. The article discusses how the language learning platform 'Edvibe' facilitates personalized learning, overcome educational inequality in rural areas during English learning process. This research aims to examine the impact of using 'Edvibe' platform on English language proficiency levels of rural secondary school students. Employing quasi experimental pre-test/post- test design, the study involved 30 middle school students. The research concludes that the platform 'Edvibe' is an effective tool for improving English teaching standards in rural areas and for increasing students' motivations. The study provides practical implications for schools and educators searching cost-effective methods to enhance language learning results.
Student well-being in AI-mediated EFL distance education
Nachida Babchikh1
Incorporating Artificial intelligence (AI) into the educational sector is altering the nature of teaching and learning profoundly. While contemporary research has increasingly focused on AI literacy and technological adaptation, students' psychological implications for AI-mediated learning is a topic, often ignored, to which we return presently. This thorny issue is particularly relevant in EFL distance educational milieus where meaningful engagement is highly dependent on students' ability to manage digitally-orientated interactions. However, research continues to rely on emotional conception developed for human-to-human interaction. As learners interact with intelligent systems through generative AI tools, automated feedback, and adaptive learning platforms, they encounter different emotional and psychological challenges related to dependency, trust, uncertainty, and human presence. This leads us to discuss the psychological well-being of learners who are increasingly involved in AI interactions. To this end, the current research highlights the importance of reconsidering the significance of understanding learners' psychological well-being in AI-enhanced learning environment. Grounded on scholarly research in distance education, educational psychology, and human-AI interaction, the paper argues that student well-being should be recognised as a central condition for effective AI integration rather than a secondary outcome of technological innovation. The study concludes by outlining implications for transferring the issue from techno-centric approaches to human-centred perspectives in distance education and highlighting future research on the psychological dimensions of educational transformation.
Challenges and recommendations.
Bouchenak Sarah1; Fathi Djerouane2
Algeria has undergone a recent educational reform at the primary school level under the Algerian President's instructions to introduce English language subject at this level starting from the third year of primary school in the school year 2022-2023. This major step leads towards promoting Algeria's position on the global linguistic landscape, and enhancing English language teaching in our country. Due to the tremendous developments taking place globally, English proficiency has become an urgent necessity in all fields and at all levels. This initiative faces several challenges in terms of methods, curricula, and human resources. Through employing a descriptive-analytical approach, this paper seeks to examine the current situation of English language teaching in Algerian primary schools in light of these recent educational reforms, and analyze the challenges hindering the success of this project, particularly the pedagogical training of teachers, the availability of resources, and the adopted curricula. This paper concludes by proposing a set of practical and well-considered recommendations that take into account the specificities of the Algerian school environment aiming to improve their quality and ensure the success and effectiveness of this reform.
Rima Belouari1; Wafa Dich1
Artificial intelligence is increasingly revolutionizing the field of English as Foreign Language. As a consequence, the use of AI tools has become a significant learning habit for the EFL students. Using them particularly outside the classroom, the EFL students generally use such tools to practice speaking, check grammar, learn vocabulary and translate expressions from the mother language to English. Such AI tools provide the students with an engaging, immersive, personalized and adaptive learning environment. This learning habit doesn't only allow the students to improve their English proficiency and fluency, but it also enhances their confidence and motivation. This study aims to investigate EFL students' use of and attitudes towards the integration of AI tools in learning English. More specifically, it examines the types of AI tools which are commonly employed, the linguistic purposes for which the students use these tools as well as their perceptions towards the effectiveness of these tools in developing their language skills. To this end, a questionnaire was administered to L1 students at Ahmed Zabana University of Relizane. The findings indicated that the use of AI tools has become an important learning practice for most of the EFL learners, employing them with different frequencies for several purposes. In addition, most of the students perceive these tools as both motivating and limiting for their critical thinking. Finally, the it was found that the students encounter some challenges when using such tools and they showed awareness toward their effective use.
Enhancing learner engagement through ethical and inclusive classroom practices
Uzma Bilal1
Artificial Intelligence is changing the way we teach English now a days. It is making learning personal and fun for students. Artificial Intelligence tools like computer programs that can talk to students, help them for writing and online learning platforms that adjust to each students needs are very useful. These tools help teachers give feedback, teach students in ways and make learning more interesting. In Cambridge classrooms, where the goals to help students communicate well think critically and learn on their own Artificial Intelligence can help teachers come up with new and exciting ways to teach. There are some problems with using Artificial Intelligence in language teaching. We need to think about things like cheating keeping student information making sure the computer programs are fair. We also do not want students to rely much on technology and must be creative in their own ways. These are issues that we need to address. Teachers have a role to play in making sure Artificial Intelligence is used in a way that is fair, transparent and helps students learn about technology. This paper looks at how Artificial Intelligence can be used in English Language Teaching in a way that's fair and inclusive. It explores ways to use Artificial Intelligence in the classroom that put the student first and make sure everyone has a chance to learn. The paper says that teachers need to use Artificial Intelligence in a way that combines technology with good teaching methods. This way Artificial Intelligence can be a tool for teachers, not a replacement for them. This discussion is part of a conversation about using Artificial Intelligence in language teaching in a responsible way. It gives teachers ideas, about how to prepare students for a world that's increasingly digital.
Code-switching fatigue and lexical gaps in AI-mediated mother tongue instruction
Pethias Siame1
Mother tongue education in rural Zambia faces a paradoxical challenge: AI tools promise scalable literacy support, but their training data exclude Mambwe (spoken in Northern Province, districts of Mbala and Mpulungu) and Lungu (around Lake Tanganyika). This paper reports on a participatory design study with 45 Grade 5 teachers and 320 learners across three zone schools, where a prototype chatbot was deployed for Mambwe reading comprehension exercises. The chatbot, fine-tuned on a small corpus of Mambwe folktales and church primers, produced two unanticipated outcomes. First, learners exhibited code-switching fatigue: the chatbot's inability to handle natural Mambwe-English alternations forced learners into artificial monolingual production, reducing spontaneous language use by 58%. Second, lexical gaps emerged in domains such as digital navigation (kupenda – to click; absent) and emotional expression (chilowola – confusion after a riddle; present but misused). The paper argues that current mother-tongue AI tools replicate colonial language ideologies by privileging standardised, written varieties over oral, heteroglossic practices. In response, the study proposes a minimally supervised lexicon expansion protocol using community radio recordings. In foreign-language English education, LLMs inadvertently penalise Lungu learners whose English shows substrate transfer (e.g., omitted articles), flagging them as errors rather than as transitional competences. The paper offers a bilingual ethical framework for AI deployment in Zambian mother-tongue classrooms.
Rethinking English language teaching in higher education
Sougand Akbarian1
Artificial intelligence (AI) is reshaping higher education by transforming how students access knowledge, engage with learning materials, and demonstrate academic competencies. Traditional language teaching approaches, often grounded in content acquisition and memorization, require reconsideration in light of AI-assisted learning environments. This study examines how AI can support competency-based English language education while fostering critical thinking, communication, collaboration, creativity, and ethical AI literacy. This conceptual study employs a comprehensive review and synthesis of recent scholarly literature on artificial intelligence, competency-based education, English language teaching (ELT), and higher education. Drawing on contemporary educational theories and emerging AI pedagogies, the paper develops an integrative framework that connects AI technologies with competency-oriented language learning outcomes. The analysis indicates that AI-powered tools can enhance personalized learning, formative assessment, academic writing support, speaking practice, and learner autonomy. However, effective implementation depends on pedagogically informed instructional design, ethical AI use, digital literacy, and the development of authentic assessment strategies that prioritize demonstrated competencies over rote knowledge reproduction. The study argues that AI should function as a pedagogical partner rather than a replacement for educators. It proposes a competency-based model that integrates AI to cultivate higher-order thinking, intercultural communication, and responsible technology use in English language education. The findings offer practical implications for educators, curriculum designers, and policymakers seeking to align language education with the evolving demands of AI-enabled higher education while maintaining academic integrity and human-centered learning.
How learner agency and AI integration shape motivation, interaction, and accuracy in LSP language learning
Tilen Smajla1
Through a survey of 513 university students in a criminal justice and security LSP course, this study investigates the effects of LLMs on language learning and instruction. Four hypotheses on motivation, interaction quality, critical assessment, and integration are tested using non-parametric techniques. Compared to peers who perceive AI as neutral or harmful, students who believe it eases learning report far greater enthusiasm and active involvement (p < 0.001). Interaction quality, prompt effectiveness, and comprehension of AI output are considerably greater among students using AI proactively and autonomously rather than responding to teacher-directed prompts (p < 0.001). No discernible difference emerged in self-reported grammatical correctness between those who critically check AI replies and those who accept them without question (p = 0.72), indicating verification alone does not ensure increased accuracy. Deeper AI integration shows a favorable correlation with overall satisfaction (rho = 0.529) and perceived efficacy (rho = 0.543), both significant at p < 0.001. These results suggest learner satisfaction and engagement in AI-supported language learning are driven by learner agency and perceived utility rather than mere access to AI technologies. Proactive, self-directed AI use may improve interaction quality and motivation more effectively than teacher-controlled, prescriptive adoption. The findings add empirical support to discussions on the pedagogical significance of LLMs while cautioning that critical assessment skills may need pairing with explicit accuracy-focused teaching to produce measurable grammatical gains. Implications for task design, evaluation, and teacher preparation in LLM-integrated courses are examined, offering curriculum designers recommendations for balancing autonomy with pedagogical scaffolding.
Interdisciplinary perspectives on AI, translation, and educational equity in Africa
Nyatundo George Oruongo1
Linguistic pluralism in Africa withers within ill-designed artificial intelligence (AI) translation systems. The digital form clothed in colonial era one-size-fits-all linguistic suit entrenches educational inequity across the African region. This paper broadly interrogates the intercourse of AI, translation and access to education in Africa. The study critically examines how today's' AI translation tools imitate colonial linguistic prejudices in African educational settings. It also lays down normative frameworks for ethically responsible AI translation advancement that nourishes African linguistic and cultural sovereignty. The study hypothesizes that European AI translation models impregnated with prejudicial colonial design principles systematically replicate and reinforce historical inequalities in African educational systems. Contribution by Canagarajah (2013) proves that language control suppresses epistemic plurality. Ngugi wa Thiong'o (1986) persuades that linguistic displacement is cultural dispossession. Mignolo (2009) pictures this within coloniality of knowledge. Nekoto et al., (2020) expose a glaring commercial underperformance on African languages. Addo et al. (2023) and the Masakhane Research Foundation document the dearth of African-language training datasets, a structural gap this paper fills. This doctrinal study finds no authoritative continental framework that governs linguistically equitable AI deployment in African education. This proposal contributes to a rights-based, culturally responsive framework for AI translation governance in African educational systems.
A bibliometric review
Nowshin Tabassum Mashaba1; Mohammad Badruddoza Talukder2
The rapid expansion of artificial intelligence in higher education has reshaped how international students engage with academic language, writing, and communication. For multilingual students, AI tools such as ChatGPT, machine translation systems, grammar checkers, and AI writing assistants have become important resources for navigating English-dominant academic environments. These tools may support students in understanding academic content, developing written assignments, translating ideas across languages, and participating more confidently in university settings. At the same time, the growing use of AI raises important questions about translanguaging, linguistic equity, authorship, academic integrity, and linguistic justice. This study conducted a bibliometric review of research on AI-mediated translanguaging and linguistic justice among international students. Using publications indexed in Scopus, the study mapped the development of this emerging research area by examining publication trends, influential authors, leading journals, contributing countries, institutional affiliations, citation patterns, and keyword networks. Bibliometric techniques, including citation analysis, co-authorship analysis, co-citation analysis, bibliographic coupling, and keyword co-occurrence analysis, were used to identify the intellectual structure and evolution of the field. The review showed how research on artificial intelligence, multilingual academic practices, translanguaging, and linguistic justice has developed over time within higher education contexts. The findings identified dominant research trends, influential sources, and emerging areas for future investigation. This study contributes to the growing discussion on AI-supported multilingual academic communication by highlighting how artificial intelligence may function both as a support system and as a site of inequality for international students.
Preserving cultural nuance in AI-generated translation
Hadil Betit1; Rima Medjedoub2
Artificial intelligence (AI) is increasingly shaping translation practices by making multilingual communication faster and more accessible. However, that rapid development raises serious concerns about whether AI-generated translation can preserve cultural nuance and context-dependent meaning that human translators commonly manage. The current paper primarily focuses on the role of AI-generated translation in translating culturally-bound expressions, identifying the ethical implications as well as the opportunities and challenges that may arise when translation is generated by machine systems that produce accurate yet often literal translations that ignore cultural nuance. To meet this objective, this paper critically synthesises twenty research studies conducted on AI-generated translation, focusing mainly on research published between 2024 and 2026 across diverse international contexts. The contribution of the paper lies in clarifying how AI-generated translation reshapes roles, practices, and expectations in the translation workflow in today's digital era, emphasising its potential to facilitate multilingual communication while also highlighting its limitations in handling cultural nuances. The synthesis of the selected studies reveals promising opportunities; including increasing accessibility, speed, and cost efficiency; enhancing real-time communication; and opening new avenues for linguistic research; as well as persistent challenges related to translating idiomatic expressions, metaphors, and context-dependent phrases. The use of AI in translation also has ethical implications, such as bias in outputs, data privacy concerns, and the risk of overdependence on automated systems. The paper concludes with a set of recommendations addressed to translators and language-technology developers for improving the use of AI in translation and suggests some directions for future research.
Opportunities, risks, and cultural implications
Rifqat Opeyemi Sanni1; Rashidat Adeola Ibrahim2
The emergence of Artificial Intelligence is reshaping language preservation and communication globally, creating new opportunities for the survival of indigenous languages. This study examines the potential of AI technologies to support the preservation and revitalization of the Yoruba language while exploring the associated cultural and linguistic challenges. Using a qualitative literature review approach, the study synthesizes existing scholarship on machine translation, speech recognition, natural language processing, and large language models in relation to indigenous language preservation. The paper investigates how effectively current AI systems represent Yoruba language structures, tonal features, literary devices, and culturally embedded meanings. It further explores the role of AI in promoting language learning, enhancing access to educational resources, increasing digital visibility, and encouraging language use among younger generations. However, the study identifies critical concerns, including language distortion, cultural misrepresentation, and the loss of contextual meanings when Yoruba expressions are translated or generated by AI systems trained predominantly on Western language datasets. Findings suggest that AI offers significant opportunities for documenting, preserving, and disseminating Yoruba language and cultural heritage through digital archives, intelligent learning platforms, and automated language tools. Nevertheless, the effectiveness of these technologies depends on the availability of culturally representative datasets and the active involvement of linguists and cultural experts in AI development. The study concludes that AI can serve as a valuable tool for Yoruba language preservation when guided by culturally responsive and ethically grounded frameworks. Collaborative efforts among AI developers, linguists, educators, and policymakers are essential to ensuring that technological innovation strengthens rather than diminishes indigenous language and cultural heritage.
V. Murali Manohar1
Language plays a vital role in education, yet linguistic differences often limit access to learning resources and meaningful participation in the teaching-learning process. Recent advances in Artificial Intelligence (AI) have created new opportunities to address these challenges through multilingual translation of text, speech, videos, presentations, and educational content. Based on a review of contemporary AI tools and emerging educational practices, this paper explores how AI-powered translation can support multilingual education and improve access to knowledge. The study highlights significant innovation in tools that not only facilitate translation but also help educators prepare lessons, organise content, engage students, and support interactive learning. As learners increasingly rely on digital technologies, educators must adopt approaches that connect students in familiar digital environments. AI tools can help teachers present content in languages that learners understand, making learning more accessible, engaging, and inclusive. At the same time, the paper recognises concerns related to translation accuracy, cultural context, ethical use, and the reliability of AI-generated content. It argues that Human Intelligence (HI) remains central to education. While AI can simplify routine tasks and support teaching practices, it cannot replace educators' judgment, empathy, creativity, and pedagogical expertise. The study concludes that AI should be viewed as a supportive partner rather than a replacement for teachers. When used responsibly and ethically, AI-powered translation can help create more inclusive, learner-centred, and effective educational experiences.
Machine translation, human agency, and the EMI educator as translator
Salemkour Imene1; Belabbes Zouleykha2
As machine translation (MT) grows increasingly fluent, the human translator is not being replaced but redefined. This study explores that redefinition within English as a Medium of Instruction (EMI), where content and language learning intersect. In EMI classrooms, lecturers and students often act as de facto translators -negotiating meaning across languages using AI-powered translation tools. Adopting a theoretical descriptive methodology, the paper aims to investigate how MT reshapes the human translator's role from sole author to critical curator. Findings revealed that when used collaboratively, MT enhances cross-linguistic comprehension, accelerates feedback, and allows human agents to focus on cultural nuance, pragmatic inference, and ethical judgment—tasks machines cannot yet own. However, risks persist: algorithmic bias, over-reliance on automated fluency, and the erosion of students' own translation skills. The study proposes a pragmatic, human-centered model: machine translation as a co-pilot, not a replacement. In EMI contexts, this means transforming translation from a hidden necessity into a visible, teachable competence. The human translator of the AI era becomes a pedagogic mediator-guiding, correcting, and culturally mediating between algorithmic outputs and learning goals. Ultimately, AI tools complement rather than replace human instruction, bridging pedagogy with digital transformation while preserving ethical responsibility
A critical discourse analysis of hedging and ideological positioning in ChatGPT
Kaouther Zerkaoui1; Kermezli Abdelkader2
Large Language Models (LLMs) are increasingly promoted as objective and impartial tools that can provide balanced information across a wide range of topics. Despite growing scholarly attention to algorithmic bias in artificial intelligence, comparatively little research has examined the linguistic mechanisms through which AI-generated discourse constructs an appearance of neutrality. This issue is particularly significant in the case of ChatGPT, whose responses are frequently perceived as unbiased despite concerns regarding the ideological assumptions embedded within its training data and output generation processes. Drawing on Norman Fairclough's three-dimensional model of Critical Discourse Analysis (CDA), this paper investigates the role of hedging as a discursive strategy in the construction of neutrality within ChatGPT responses. The study analyzes a corpus of responses generated from prompts addressing selected controversial social issues, with particular attention to linguistic features such as modality, mitigation, evidentiality, attribution, and evaluative language. Through a systematic examination of expressions such as "some argue," "it may be," and "research suggests," the paper explores how ChatGPT negotiates competing viewpoints while simultaneously positioning certain perspectives as more legitimate, credible, or authoritative than others. The study argues that neutrality in AI-generated discourse should not be understood as the absence of ideology, but rather as a discursively constructed effect achieved through specific linguistic choices. By foregrounding hedging as a central mechanism in the production of apparent objectivity, this research contributes to emerging discussions on language, ideology, and algorithmic meaning-making. More broadly, it demonstrates the relevance of Critical Discourse Analysis for understanding the social and epistemological implications of generative artificial intelligence in contemporary digital communication.
Risks of misrepresentation in Indian regional media
Shyamali Banerjee1; Suhas S Kashyap2
Given the context that there is a subtle difference between translation and adaptation which only lies in the literal meaning of both the terms, translation is a linguistic and social-political construct, whereas adaptation involves the reconstruction of all the above three factors. The concept of adaptation or translation from one art form to another, like literature to film and theater and vice versa or from one language to another, has always remained problematic because of its inherent nuances like essence of originality, auteur signorial motifs or references, narrative digression and other technical differences in terms of representation or misrepresentation of socio-political or cultural context. In the light of intellectual authorship, originality, creativity, the translation may create wither standardization or differentiation of the text. Hence, Translation can be defined as the offshoot of a linguistic-textual intersectional mechanism in which a text in one language is re-contextualized in another language. As a linguistic textual operation, translation is, however, subject to, and substantially influenced by, a variety of extra-linguistic factors and conditions. It is this interaction between 'inner' linguistic-textual and 'outer' extra-linguistic, contextual factors that makes translation such a complex phenomenon. The rapid integration of Artificial Intelligence (AI) in translation, particularly through systems such as Google Translate and ChatGPT, has transformed the landscape of journalism in multilingual societies like India. Regional media content is increasingly translated into English or other dominant languages for national and global dissemination. While this enhances accessibility and reach, it introduces significant risks of semantic distortion, cultural misrepresentation, and political bias. This paper critically examines the role of AI-mediated translation in Indian vernacular journalism, focusing on political reporting. It explores how translation inaccuracies can reshape narratives, influence public opinion, and affect democratic discourse.
Hanin Shougary1
Advances in artificial intelligence are enhancing translation practices. In such a fast-paced era, the demand for rapid information exchange is on the rise. At this point, such tools are of critical importance, especially AI-powered translation tools, which enable intellectuals to work across languages. This study explores how AI-powered translation improves efficiency, consistency, and accessibility. AI-based translation systems significantly reduce the time required to perform academic tasks due to their immediate responses. This substantially enhances the overall productivity in academic workflows. Additionally, in specific fields of expertise, accuracy is vital to ensure consistency in terminology. This leads to a uniformity of terms in different texts, maintaining coherence in academic discourse. The reason why this is particularly important is that even minor inconsistencies can lead to misinterpretations and misapprehensions. Furthermore, AI tools are considered important devices that assist non-native English speakers in scholarly communication and in their contributions to international academic communities. This reflects enhanced accessibility to academic content by enabling users to engage with and understand scholarly materials regardless of linguistic proficiency. AI is defined as an assistant that facilitates the translation process and produces clearer and more structured texts, rather than a substitute for human translators. AI-assisted translation represents a significant development in applied linguistics, particularly in the field of English for Academic Purposes, where accuracy and clarity are crucial. Thus, the act of integrating artificial intelligence in academic contexts leads to effective communication and broader access to knowledge, asserting the notable contributions of artificial intelligence in translation procedures.
A computational pragmatic-linguistic approach
Nisrine El Hannach1
This study examines the phenomenon of hallucination in large language models from a computational linguistic perspective, moving beyond its reduction to a merely technical or informational error, and recontextualizing it within the broader problem of ambiguity and transformations in meaning production. While ambiguity in traditional linguistic theory is understood as a legitimate multiplicity of interpretive possibilities resulting from lexical, syntactic, pragmatic, or referential indeterminacy, hallucination reveals a different form of ambiguity—one that does not arise from multiple possible meanings, but from the generation of meaning lacking textual, referential, or contextual grounding. Accordingly, the study raises a central question: does hallucination represent a contemporary extension of the concept of ambiguity, or does it constitute a conceptual rupture from it? The study argues that hallucination is not merely a statistical deviation in model performance, but rather a form of generative ambiguity resulting from the dissociation of linguistic production from the pragmatic and referential constraints that regulate meaning in natural communication. In this process, probability shifts from being a mechanism for interpreting meaning to becoming a mechanism for producing semantic surplus that may ultimately lead to fabrication. Based on this premise, the study proposes a reconceptualization of hallucination through its association with the notion of Overgeneration of Meaning, and develops a computational linguistic classification of its patterns, including referential, semantic, pragmatic, and inferential hallucinations. Methodologically, the study adopts an approach that combines theoretical analysis with applied investigation, drawing on semantic linguistics, pragmatics, and computational linguistics. It also employs applied samples drawn from the outputs of large language models in order to identify hallucination patterns and analyze their mechanisms in light of the concepts of ambiguity, reference, relevance, and inference. The study concludes that hallucination should not be understood merely as a reliability issue in large language models, but rather as evidence of a deeper transformation in the concept of ambiguity itself, thereby calling for an expanded theoretical framework for the study of meaning in the age of generative artificial intelligence
Reproducing Global linguistic inequality in multilingual knowledge production
Widjdane Slimani1; Nour Elhouda Remil2
Generative artificial intelligence has been widely presented as a democratizing force capable of expanding access to knowledge across linguistic and geographical boundaries. However, emerging debates suggest that these systems may simultaneously reproduce existing global inequalities through unequal language representation, data availability, and performance disparities across linguistic communities. This paper critically examines how generative AI technologies may contribute to the formation of an "invisible language hierarchy," in which dominant languages, particularly English acquire greater visibility, usability, and epistemic authority than less digitally represented languages. Drawing on scholarship in digital inequality, language ideology, and sociolinguistics, the paper explores how the architecture and training conditions of large language models influence multilingual knowledge production and access. Particular attention is given to "English, Arabic, French" dynamics within the MENA region, where multilingual users frequently navigate educational, professional, and cultural spaces through unequal linguistic infrastructures. The discussion considers how differences in training data volume, linguistic resources, and model optimization may affect translation quality, content generation, cultural nuance, and access to educational opportunities. Rather than approaching AI as a neutral technological tool, this paper argues that language technologies participate in broader social and epistemic structures that shape whose voices are amplified and whose knowledge remains peripheral. The paper concludes by proposing principles for more linguistically inclusive and socially responsible AI development in education and public knowledge environments.
Multilingualism as an epistemic challenge to AI-assisted legal discourse analysis
Amani Maboud1; Rima Medjedoub2
Artificial intelligence tools for linguistic analysis have advanced considerably in their capacity to detect patterns and describe language use at scale. Yet this progress has unfolded largely within monolingual, English-dominant contexts, leaving AI instruments poorly equipped for multilingual legal discourse. This paper argues that multilingualism is not merely a technical obstacle to AI-assisted analysis but a fundamental epistemic challenge that exposes the limits of computational approaches and the enduring necessity of context-sensitive human interpretation. The Algerian sociolinguistic setting serves as the theoretically productive case, given the complex coexistence of Modern Standard Arabic, Algerian Darija, Tamazight, and French within a single legal-institutional space. Drawing on forensic linguistics, corpus-assisted discourse studies, language ideology theory, and NLP resource literature, the paper advances a three-dimensional argument. First, code-switching in Algerian legal discourse carries institutionally motivated indexical meaning that AI tools cannot recover without interpretive mediation. Second, monolingual training data structurally reproduces the NLP-humanities gap, embedding epistemic asymmetries into the analytical infrastructure itself. Third, a forensic sociolinguistics of multilingual institutional discourse demands principled methodological complementarity between computational and human modes of analysis, not as a provisional workaround but as a coherent research framework. The paper concludes that methodological transparency, contextual accountability, and sociolinguistic sensitivity are necessary foundations of responsible AI-assisted analysis, not optional refinements. Recommendations are offered for researchers in comparable under-resourced contexts, including prioritizing open-source tools and developing domain-specific annotated corpora for dialectal legal registers.
Code switching in human-AI interactions in ChatGPT discourse.
Meryem Selka1
Within multilingual speech communities, the use of more than one code in a single interaction is a completely predictable behavior. Human beings tend to fully use their linguistic force in daily interactions. As artificial intelligence (AI) has settled among the community, serving people for the smallest tasks, the inquiry on whether such tools can handle code switching (CS) as naturally as humans do raises. The current study analyses human-AI interactions all containing instances of CS through a sociolinguistic lens. The intended goal, revealing whether AI tools such as ChatGPT are primarily monolingual tools, gravitating towards English as a dominant language or if they do accommodate to the use of CS when the users trigger a mixture of linguistic varieties in a conversation. Moreover, the research seeks to identify any noticeable patters used by the AI tool when it comes to CS. To achieve this end, a number of corpora featuring human-AI CS interactions were meticulously analyzed. The used corpora included human-AI interactions involving prompts including switches between English, French and Arabic. The results of the study show that ChatGPT switches to another code, attempting to match the patterns used by the human party of the interaction but only to a certain extent. The AI tool may preserve a monolingual approach by reducing the multilingual switches and offering translations. This proves that the tool is primarily monolingual but would occasionally CS solely if the switch was triggered by the interlocutor.
Cultural-linguistic bias in large language models and the erasure of Arabic-Francophone EFL learners
Moufida Boumous1
Large language models (LLMs) have rapidly entered EFL classrooms worldwide as writing assistants, feedback tools, and language learning companions. Yet the pedagogical discourse surrounding their adoption overwhelmingly assumes a learner shaped by Anglophone, monolingual, and Western educational norms. This paper critically examines the cultural-linguistic biases embedded in dominant LLMs and argues that these biases systematically disadvantage Arabic-Francophone EFL learners, particularly those situated in North African institutional contexts. Drawing on postcolonial applied linguistics, critical AI studies, and the author's sustained classroom experience in Algeria, the paper identifies three interlocking dimensions of bias. First, LLMs trained predominantly on English-dominant Western corpora pathologise rhetorical and pragmatic conventions rooted in Arabic and Francophone academic traditions, recoding culturally valid communicative choices as errors. Second, cross-linguistic transfer patterns common among trilingual Arabic-French-English learners are rendered invisible by models that lack any metalinguistic awareness of the learner's actual linguistic ecology. Third, when students systematically comply with LLM feedback, they undergo a process of linguistic ventriloquism: their writing becomes algorithmically approved yet pragmatically hollowed, severed from the communicative identities that sustain it. The paper argues that Algerian and broader North African EFL contexts are not peripheral case studies but diagnostic sites that expose the ideological architecture undergirding mainstream AI-language research. A postcolonial reorientation of this field is not merely desirable; it is a condition of intellectual honesty. The paper concludes by proposing critical AI literacy as a pedagogical imperative for EFL instructors operating outside the Anglophone centre.
A sociolinguistic and ethical analysis.
Kamel Eddine Atallah1; Latifa Amerani1
Linguistic Bias in Artificial Intelligence: A Sociolinguistic and Ethical Analysis The rapid integration of Artificial Intelligence (AI) into communication, education, translation, and information systems has transformed the way language is produced, interpreted, and disseminated. Despite its technological advancements, AI systems often reproduce and amplify existing linguistic inequalities embedded in the data on which they are trained. This study examines linguistic bias in AI from a sociolinguistic and ethical perspective, focusing on how language varieties, dialects, minority languages, and culturally specific forms of expression are represented within contemporary AI technologies. Drawing on sociolinguistic theories of language variation, language ideology, and linguistic power, the paper explores the mechanisms through which AI privileges dominant languages and standardized forms while marginalizing less-resourced linguistic communities. The study further investigates the ethical implications of linguistic bias, including issues of fairness, inclusivity, cultural representation, and digital linguistic rights. Through an analysis of AI-driven applications such as machine translation systems, conversational agents, and automated language assessment tools, the paper highlights how linguistic bias can influence access to information, educational opportunities, and social participation. The findings suggest that linguistic bias is not merely a technical problem but also a social and ethical challenge that reflects broader structures of power and inequality. The paper argues for the development of more linguistically inclusive AI systems through interdisciplinary collaboration among computer scientists, linguists, educators, policymakers, and ethicists. By promoting greater linguistic diversity in training data and adopting ethical frameworks that recognize the value of all language varieties, AI technologies can contribute to a more equitable and socially responsible digital future. This research contributes to ongoing discussions on artificial intelligence, language, and social justice by emphasizing the need for culturally sensitive and linguistically inclusive technological innovation.
AI and the challenge of translating poetry
Nouha Abbas Chohra1; Fethia Braik2
Recent developments in Artificial Intelligence (AI) have marked a quantum leap in the fields of language learning and literary analysis, offering increasingly sophisticated tools for reading, understanding, and interpreting literary texts. As AI tools become more integrated into academic and educational environments, evaluating their ability to analyze complex literary works particularly poetry becomes both strategic and vital. This paper explores the intersection of artificial intelligence, literary interpretation, and translation, with a focus on the challenges AI faces when attempting to process poetic language.Due to its intricate interplay of metaphor, emotional subtlety, symbolic ambiguity, and rhythm, poetry serves as a rigorous test case for such exploration. As a central case study, Sylvia Plath's Ariel is examined for its dense imagery, psychological intensity, and feminist subtext, offering a rich terrain for automated interpretive reading. Using a generative AI model (ChatGPT), selected poems underwent digital analysis to evaluate how AI approaches meaning, voice, and emotional tone.The AI's output was critically compared with established literary interpretations and theoretical frameworks. Particular attention was paid to how AI deciphered metaphorical language, shifts in tone, and recurring themes such as rebirth, subjugation, and female agency. While the model demonstrated an ability to perform lexical analysis and identify thematic patterns, it often failed to register deeper emotional transitions or to contextualize the text within its historical, biographical, and cultural frameworks.This study also reflects on the implications of AI literary analysis for translation, particularly the translation of poetry. Given that poetic translation demands a sensitivity to emotional nuance, figurative language, and cultural codes, the shortcomings of AI in literary interpretation signal a similar limitation in its translational potential. Although AI may support human translators in identifying stylistic patterns and lexical parallels, it remains constrained in its ability to render the layered meanings, rhythms, and symbolic depths that define poetic texts.Ultimately, this paper argues that while AI can assist in certain analytical or preparatory aspects of literary and poetic translation, it cannot replace the human insight, creativity, and cultural literacy necessary for faithful and nuanced translation of poetic works.
Evidence from Algerian EMI biology classrooms
Abdelghani Mim1; Abdellah Bouguenous2
The rapid expansion of machine translation tools is reshaping linguistic practices in higher education. While research has examined translanguaging and first-language use in multilingual classrooms, spontaneous translation as an autonomous cognitive strategy in English-Medium Instruction (EMI) contexts remains underexplored, particularly in Global South settings such as Algeria, where Arabic, French, and English coexist. This paper investigates how Algerian university students enrolled in EMI biology programmes spontaneously employ machine translation to facilitate comprehension and language learning, with particular attention to its perceived contribution to grammatical competence. Adopting an exploratory descriptive case study design, the research was conducted at the University of El Oued (Algeria) and involved 100 EMI biology students. Data were collected through questionnaires and semi-structured interviews to explore students' machine translation practices, perceptions of their grammatical development, and views on the growing use of GenAI tools. The study addresses the following research question: How do EMI biology students perceive the role of spontaneous translation in supporting their grammatical competence within a trilingual context (Arabic–French–English)?The analysis is informed by translation studies and translanguaging perspectives. Findings suggest that students perceive spontaneous translation as facilitating grammatical awareness through structural comparisons between Arabic, French, and English, enabling them to notice linguistic patterns, clarify grammatical relationships, and support comprehension. Participants also expressed concerns that excessive reliance on AI translation tools may reduce opportunities for active linguistic processing and reflective language learning. The study highlights the continued pedagogical relevance of learner-generated translation strategies and the responsible integration of AI technologies in linguistically diverse higher education contexts.
A quality comparative study .
Slimane Fatma1
Translation has long been considered a deeply human, cognitive skill. However, the prevalent and rapid advancement of AI , neural machine translation to be more precise , and large language models has fundamentally challenged this assumption making us question the nature ,standards ,and future of translation quality . This paper offers a theoretical examination of the quality dimensions that distinguish human translation from machine translation without relying on empirical experimentation .Based on established theoretical frameworks _ including the multidimensional quality metrics model , Juliane House's translation quality assessment model ,and Eugene Nida's concepts of formal and dynamic equivalence _the study crucially , analyses the core parameters by which translation quality is defined and measured . The paper argues that despite the remarkable progress of machine translation in linguistic accuracy and fluency ,particularly in high source language pairs ,it failed in other areas or dimensions that necessitate deeper cognitive and cultural competence ,specifically : pragmatic coherence ,stylistic sensitivity ,cultural mediation ,and contextual judgment .These remain ,at present ,distinctly human strengths rooted in lived experience ,intentionality ,and intercultural awareness .Furthermore ,the paper explores the implications of this theoretical comparison for language teaching and translator training ,proposing that educators must equip learners not only with translation competence ,but also with the critical ability to evaluate ,interrogate ,and judiciously post –edit AI generated translations .The study concludes that human and machine translation not conflicting modalities ,but rather complementary .Their productive integration depends on a clear theoretical understanding of what each can _and can't do .
A perspective from cognitive linguistics
Pham Huu Duc1
The fast development of artificial intelligence (AI) has helped language learners to use AI Bots to help them with translation work, but simultaneously but has simultaneously made them depend heavily on the tools, and the translation quality remains inconsistent due to linguistic, cultural, and contextual disparities. From the perspective of cognitive linguistics, language learners need to incorporate conceptual blending and metaphorical mappings across languages using intralingual rewording or interlingual rephrasing. Cognitive linguistics focuses on the meaning coming from the interaction between language, thought, and cultural context, while AI Bots rely solely on vast corpora and modeling to interpret language structures provided. The study investigates how language learners use AI Bots in helping with translation work. This study presents the use of AI translation in learning translation from the perspective of cognitive linguistics. The study examines translation samples to enhance translation quality through a corpus-based approach. A parallel corpus of English - Vietnamese sentence pairs are constructed, covering such domains as business, media, and general language. The data are analyzed using two translation strategies: AI Translation (AIT) and Human Translation (HT), grounded in such key theoretical frameworks as the theory of conceptual metaphor (Kövecses, 2005; Lakoff & Johnson, 1999, 2008) and the theory of semantic frame (Fillmore (1976, 1982; Croft & Cruse, 2004). The features of cognitive linguistics are displayed that are not characteristic of AI Bots and present how language learners acquire cognitive translation of languages. The study aims to demonstrate that translation works with the assistance of theoretical frameworks will serve as a systematic decision-making framework that enhances semantic accuracy, fluency, and communicative effectiveness in different languages. The findings reveal a substantial improvement in translation quality when theoretical frameworks are applied, with the number of errors reduced. These findings display the importance of integrating cognitive theories into both translation teaching and translator training. This suggests the need for combining theoretical knowledge with technological tools in the evolving landscape of translation studies.
A literature review on who AI works best for
Chiraz Retimi1
As AI language tools become central to everyday communication, a critical question emerges: do these systems work equally well for all users? This paper presents a systematic literature review arguing that current AI speech recognition and text processing technologies are not linguistically neutral. They systematically favor certain speaker profiles while disadvantaging others. Drawing on sociolinguistics, psycholinguistics, and human-AI interaction research, this study examines the key dimensions along which AI performance varies: nativeness, age, gender, dialect, language resource availability, and socioeconomic background. Studies consistently show that AI tools trained predominantly on data from young, educated, native speakers of high-resource languages underperform for users outside this profile. Non-native speakers, elderly users, dialect speakers, and speakers of low-resource languages such as Arabic and Amazigh encounter higher error rates, miscommunication, and unequal access to AI-mediated communication. This review synthesizes existing findings across these dimensions, identifies gaps in current research, and calls for a more inclusive approach to AI language tool development. The findings have direct implications for developers, educators, and policy makers advocating for linguistic justice in AI design.
Algorithmic standardization and linguistic hierarchies in AI-mediated multilingual communication
Sarra Soumia Boudjadi1
The multilingual interaction landscape is currently witnessing a rapid proliferation of generative AI platforms. Such technologies are reshaping identity construction across generations while expanding communicative access and facilitating cross-linguistic interaction with efficiency, accessibility, and speed. Notwithstanding the undeniable merits, these tools appear to encourage forms of linguistic uniformity that are often subtle enough to escape immediate scrutiny. Contextually negotiated meaning, indexical expressions, and regional inflections are gradually absorbed in this process. Yet, relatively little scholarly attention has been paid to the sociolinguistic consequences of this algorithmic standardization. Therefore, the present study aims at examining how AI systems consistently privilege standardized varieties associated with global intelligibility, thus reinforcing dominant linguistic hierarchies while diminishing the expressive richness associated with regional, minority, and non-prestige language varieties. Grounded in Blommaert's (2010) sociolinguistics of globalization, language ideology theory (Woolard & Schieffelin, 1994), and Fairclough's (2003) framework of Critical Discourse Analysis (CDA), the study adopts a mixed-methods approach. It analyzes 50 purposively sampled expressions drawn from two languages that occupy structurally asymmetrical positions within global linguistic hierarchies: Arabic and English, processed through generative AI systems including ChatGPT, Gemini, and machine translation tools such as Google Translate and DeepL Translator to foreground dialectal, indexical, and code-switching features. Quantitative frequency analysis identifies patterns of lexical simplification and dialectal neutralization while qualitative CDA interprets their sociolinguistic and ideological implications. Accordingly, it advocates for a more linguistically equitable approach that actively sustains nuance and preserves the sociolinguistic plurality characterising real-world multilingual communication.
Where languages meet, adapt, and thrive
Nihad Albeik1
The Multilingual Mind: Where Languages Meet, Adapt, and Thrive explores multilingualism as a dynamic cognitive, cultural, and social phenomenon rather than a simple accumulation of separate linguistic systems. The chapter examines how multiple languages coexist within the mind, interact with one another, and shape patterns of thought, identity, memory, communication, and learning. It highlights the multilingual mind as a flexible space where languages do not merely compete, but constantly influence, support, and reshape each other through code-switching, transfer, translation, and cultural negotiation. The discussion also considers how multilingual individuals navigate different emotional worlds, social expectations, and communicative contexts, developing adaptive strategies that enrich both personal expression and intercultural understanding. It further considers what these multilingual capacities mean for artificial intelligence in education, culture, and society, including how AI-based tools for translation, tutoring, and communication handle code-switching, emotional register, and culturally embedded meaning. By bringing together insights from linguistics, education, psychology, and cultural studies, the chapter presents multilingualism as a living process of mental adaptation and human connection. In doing so, it raises questions about whose languages and varieties are represented in AI systems, and how AI can support, rather than flatten, the full linguistic repertoire of multilingual learners and speakers. Ultimately, it argues that multilingual minds reveal the creative capacity of language to expand identity, deepen empathy, and sustain meaningful participation in an interconnected world. As AI increasingly mediates how people learn, translate, and communicate, this perspective offers a human-centred lens for designing and using AI that respects linguistic and cultural diversity.
A usage-based and sociopragmatic perspective
Dumelang Thomas-Kereeditse1
The increasing integration of artificial intelligence (AI) in language learning has led to a growing reliance on machine translation tools. While such technologies offer immediate access to multilingual equivalents, their effectiveness in the context of under-resourced languages remains questionable. Drawing on observation and discourse analysis, this study demonstrates that learners of Setswana, which is an under-resourced language, are likely to encounter translation inaccuracies, particularly at the level of discourse. These errors arise from the inability of AI systems to adequately capture context-dependent meaning, cultural nuance, and pragmatic intent. This study employs AI-assisted translation within a Setswana proficiency classroom to evaluate its pedagogical potential and limitations. Framed by usage-based theory and sociopragmatic theory, the research examines how AI-generated target texts interact with the socially nature of meaning in language use. Findings indicate that while AI tools can support vocabulary acquisition, they fail to account for discourse-level features such as coherence, politeness strategies, and culturally specific expressions. Therefore, this study argues that AI translation should be used as a supplementary resource rather than a primary learning tool in Setswana language proficiency lessons. This highlights the need for critical engagement with AI outputs and emphasises the role of teacher in mediating meaning and guiding learners toward contextually appropriate language use. This contributes to broader discussion on technology-enhanced language learning in low-resource contexts.
Darinka Marolova1; Dragana Kuzmanovska2; Ana Vitanova Ringaceva2; Biljana Ivanova2
Artificial intelligence (AI) is increasingly transforming language education by offering innovative opportunities for developing students' linguistic proficiency, translation competence, and intercultural awareness. This study investigates how AI-based tools can be integrated into translation pedagogy to support the development of these competencies among German and English language students at the Faculty of Philology, Goce Delcev University in Stip, North Macedonia. The research is based on an experimental task in which students translated selected passages of German and English literary prose into Macedonian. The participants were divided into two groups: one used AI-assisted translation tools (such as ChatGPT, DeepL, and Google Translate), while the other completed the task without technological support. This design enabled a comparative analysis of students' translation processes and outcomes. The study evaluates the linguistic accuracy, stylistic quality, and cultural sensitivity of students' translations, with particular attention to interpret and transfer cultural meanings embedded in the texts. The findings indicate that students using AI tools generally produced more lexically varied and syntactically fluent translations. However, their reliance on AI sometimes resulted in the uncritical adoption of inappropriate or culturally insensitive expressions. In contrast, students working without AI demonstrated stronger interpretative engagement and greater originality, although their translations were occasionally constrained by linguistic limitations. The results suggest that AI can enhance students' translation performance and vocabulary development, but their pedagogical value depends on guided and reflective use. When integrated critically into instruction, AI-assisted translation supports the development of intercultural competence and stylistic awareness as key components of translation training.
AI ethics, algorithmic bias, and the human cost of accent reduction technology
Ana Jovic1
AI-powered language technologies are increasingly used in informal language learning, yet their ethical implications remain under-examined. Accent reduction apps, in particular, claim to empower multilingual speakers while potentially reinforcing linguistic discrimination. This paper aims to identify the ethical failures embedded in AI-powered accent reduction technology, how these failures manifest in the user experience, and the implications for AI ethics and social responsibility in education. The study is grounded in AI ethics (transparency, accountability, fairness) and critical sociolinguistics (native-speakerism, language ideology). Drawing on three months of autoethnographic engagement with BoldVoice, the analysis identifies three interconnected ethical failures: opaque accountability, the commodification of insecurity, and the individualization of structural bias, in which users are positioned as responsible for fixing a systemic problem. The paper calls for ethical accountability in AI language technologies, transparency, pluricentric design, and a commitment to social responsibility. Accent reduction apps must be interrogated not as neutral pedagogical tools but as technologies that can reinforce linguistic discrimination if left unchecked.
How large language models misread locative and aspectual markers in bantu languages
Pethias Siame1
This paper investigates the friction between Large Language Models (LLMs) and the morphosyntactic structures of Icibemba (M42), a dominant Bantu language of northern Zambia. While LLMs excel in Germanic and Romance languages, their transformer-based attention mechanisms struggle with Bantu noun class systems and agglutinative verbal morphology. Using a dataset of 1,200 Icibemba sentences annotated for locative applicatives and progressive aspectual markers, we demonstrate that GPT-4–derived models systematically overgeneralise subject-verb-object patterns, producing grammatically impossible predictions (e.g., misplacing the locative prefix ku- as a prepositional object). The study further contrasts LLM output with classroom performance of L1 Bemba-speaking primary learners in Mpulungu District, revealing that models fail to replicate child-like acquisition of chi-/fi- class agreement. Pedagogically, the paper argues against uncritical adoption of LLM-based tutoring for Bantu-language literacy and proposes a hybrid model: human-led construction grammar instruction augmented by token-level visualisations of model uncertainty. The paper concludes that without Bantu-specific training corpora and attention-head modifications, LLMs risk reinforcing standard language ideologies that erase rural Bemba, Mambwe, and Lungu varieties. This has direct implications for Zambia's language-in-education policy, which recognises seven local languages but provides no digital infrastructure for their computational linguistics.
Acoustic bias and the erasure of tone in a Lake Tanganyika language
Pethias Siame1
AI models claim linguistic neutrality, but this paper demonstrates how text-to-speech (TTS) systems systematically erase Lungu (a threatened Bantu language of Mpulungu District, with an estimated 55,000 speakers) by imposing non-tonal prosodic templates. Lungu has two contrastive tones (high and low) that distinguish lexical meaning (e.g., úmúntu 'person' vs. ùmúntu 'shadow/spirit'). Using acoustic analysis of Lungu speech from 20 native speakers (aged 55–82) and comparing it with synthetic speech from three commercial TTS models trained on "general African" corpora, it was observed that models neutralise tone into pitch-accent patterns modelled on non-tonal lingua francas. This is not a data-sparsity problem alone: even when fine-tuned on 15 hours of Lungu radio broadcasts, models fail to realise tone sandhi across word boundaries, producing semantically unpredictable outputs. The study identifies three layers of bias: (1) corpus bias - existing Lungu digital texts use orthographies that mark tone inconsistently or not at all; (2) algorithmic bias - attention mechanisms favour sequential predictability over suprasegmental features; (3) evaluation bias - standard TTS metrics (MOS, WER) treat tonal errors as minor prosodic deviations. The paper proposes a tonologically anchored evaluation suite for Lungu and adjacent Mambwe varieties. It can be argued that assuming linguistic neutrality in AI is a form of epistemic violence against tonal Bantu languages, perpetuating their exclusion from digital futures.
Educational implications of professional translators' perspectives
Majda Glavaš Rankov1; Ana Simić2
Artificial intelligence (AI) is transforming translation practice and redefining the competencies expected of professional translators, raising important questions about whether current translator education adequately prepares students for an AI-assisted profession. Building upon the findings of previous research on professional translators' attitudes toward AI, this presentation discusses the educational implications of those findings. The presentation draws on data collected from 151 professional translators in Serbia through an online questionnaire. Particular attention is devoted to findings concerning changing professional competencies, willingness to participate in AI and post-editing training, attitudes towards integrating AI-related content into translation programs, expectations regarding the future role of translators, and qualitative reflections provided in open-ended responses. The findings indicate that professional translators increasingly recognize AI literacy, post-editing, critical evaluation of machine-generated translations and lifelong learning as essential competencies for future professionals, while continuing to emphasize the importance of linguistic, cultural and translation expertise. The qualitative responses further highlight the need for adaptability and continuous professional development in response to rapid technological change. By interpreting professional translators' perspectives through an educational lens, the findings indicate the need for improving existing curricula by integrating AI literacy, post-editing, ethical AI use and critical evaluation.
Between imitation and innovation
Belahmar Louazani Nihad Soumia1
The rapid development of generative artificial intelligence (AI) has introduced new possibilities for the production, interpretation, and circulation of literary language. AI systems are increasingly capable of generating poems, narratives, dialogues, and other forms of creative writing that can resemble human-authored texts in style, structure, and emotional expression. This development raises important questions concerning the nature of literary creativity and the boundaries between imitation and innovation. While AI-generated texts often draw upon patterns and conventions derived from existing literary works, their capacity to recombine linguistic and stylistic elements challenges conventional understandings of originality and authorship. This paper explores the relationship between imitation and innovation in AI-generated literary language, focusing particularly on the concepts of creativity, originality, and authorship. It examines whether the apparent creativity of generative AI represents genuine innovation or sophisticated linguistic recombination based on previously existing human-generated material. The paper further considers how the involvement of AI complicates the traditional distinction between author and tool, particularly when literary texts are produced through collaboration between human prompts and machine-generated language. By approaching AI-generated literature as a developing form of literary production, the study highlights both its creative potential and its limitations. Ultimately, the paper argues that AI does not simply reproduce literary language but participates in transforming how literary creativity and authorship are understood. The emergence of AI-generated literature therefore calls for a reconsideration of originality, innovation, and the human role in literary creation.
Generative AI, folktales, and the flattening of oral heritage
Lina Benzerara1; Salim Kerboua2; Zakarya Senouci3
Generative AI is increasingly becoming a mediator through which cultural narratives are retold, adapted, and circulated, raising questions about how culturally specific forms of meaning survive algorithmic reconstruction. This study examines AI-generated retellings of three Algerian folktales representing distinct narrative traditions: Lalla Loundja, Bechkerker et l'Ogre, and Uccen d Insi. The tales are approached as culturally embedded narratives whose distinctive meanings emerge through different configurations of female agency, grotesque survival, animal symbolism, and pragmatic morality. Using qualitative comparative textual analysis, documented versions of the three tales are compared with retellings generated by ChatGPT through a deliberately minimal prompting protocol. The analysis examines transformations in narrative structure, character agency, moral logic, humor, symbolism, cultural attribution, and oral storytelling features. The findings reveal four recurring forms of transformation: narrative domestication, in which culturally specific conflicts are reshaped according to familiar narrative conventions; narrative compression, in which episodic and cumulative oral structures are substantially reduced; narrative substitution, in which a documented tale is replaced by a substantially different narrative; and cultural misattribution, in which regional identity is inaccurately assigned. Taken together, these transformations suggest that generative AI tends to privilege familiar plot structures, coherent moral frameworks, and accessible narrative conventions over culturally specific forms of storytelling. The study argues that AI-generated retellings constitute acts of cultural mediation rather than neutral reproduction, contributing to a process of algorithmic narrative normalization through which underrepresented oral traditions may be reshaped as they enter digital circulation.
How anisongs foster Japanese language engagement beyond translation
Yuka Shiratsuchi1; Shinji Okumura1
Anime theme songs (anisongs) serve as a major point of contact through which non-Japanese speakers engage with the Japanese language and culture. A previous qualitative study analyzing approximately 800 YouTube comments on five globally prominent anisongs proposed a "Multilayer Engagement Model," suggesting that non-Japanese speakers' engagement unfolds across four layers: acoustic, affective, symbolic, and learning. However, as this is an exploratory study based on a limited dataset, large-scale empirical validation is required to establish the model's validity and generalizability. Drawing on Iwabuchi's (2002) concept of cultural odorlessness as a theoretical framework, this study examines a large-scale corpus of YouTube comments on multiple anisongs spanning the 1990s to the 2020s. Employing a mixed-methods approach that combines AI-assisted text mining with qualitative analysis, the study aims to verify and refine the Multilayer Engagement Model. The quantitative strand involves AI-assisted text classification, frequency analysis, and correlation analysis with song attributes. The qualitative strand involves close reading of representative comments to explore the relationship between cultural engagement and Japanese language engagement. A cross-temporal comparison further examines how AI-mediated digital media environments and evolving fandom cultures have shaped patterns of engagement across decades. This study seeks to empirically clarify the mechanisms through which anisongs function as cultural catalysts independent of translation in the age of AI, offering new perspectives for research on cultural globalization and language learning motivation.
AI, colonial extraction, and black cultural production in Ryan Coogler's Sinners and Black Panther
Ivana Aladić1
The development of Artificial Intelligence has led to intense debates over ownership, authorship, the process of creation, and the ethical implications of culturally influenced creation. These technologies are heavily reliant on extraction. Generative AI is trained on cultural, linguistic, and artistic materials, often without the consent of the creators it's built on. Therefore, AI does not only create; it extracts, consumes, and appropriates the original. My main argument is that AI production shares similar qualities to colonialisation. Both processes share the dependence on the extraction and appropriation of human resources. Colonialisation aimed for territorial expansion through the appropriation of labour, goods, land, and culture. AI commodifies language, artistic production, and creative processes, erasing the source identity. Both processes are based on erasure of the source, exploitation of human labour, and its reproduction for capitalistic purposes. By analysing the work of Ryan Coogler, his movies Black Panther (2018) and Sinners (2025), this paper displays how Black tradition, creativity, and voice become prey to extraction, proprietorship, and marketable consumption. Both movies shed light on violence embedded in systems of power, built on colonial ideas, which aim to replicate, consume, and discard the communities and individuals they are stealing from. Much like these systems, AI production replicates colonial accumulation, consuming language, art, history, style, and overall identity of the source. In this aspect, the 'hunger' of AI is not solely metaphorical, but serves as a parallel to imperial structures, transporting them into the digital universe.
Borderlines in Aidan Marchine's Death of an Author
Rasha Saeed Abdullah Badurais1
This study aims at investigating Aidan Marchine's novella, Death of An Author, from a posthumanist deconstructive perspective. This goal is achieved through analyzing the unstable moments of borderlines that undo the distinction between binaries: human/ machine or technology, life/ death, and originality/ imitation. Contemporary posthumanist theory reconceptualizes frontiers and margins of being (Balo, 2025) as productive, hybrid, and relational spaces that destabilize the human/non-human divide. Derrida's concept of the borderline (1979) provides the philosophical grounding for this shift, revealing that these spaces are not external to being but are its very condition of possibility. Within these deconstructive borderlines of posthumanist identity, the article probes the novella's paratextual elements: the title, chapter 9 subtitle, chapter 12 as the afterword; along with the selected textual components: the major character, Peggy Firmin, who is the fictional author 'partially' indicated in the title, and Gus Dupin, her deconstructive trace in the text. The analysis and discussion show that the perpetual play between the above binaries leads to highlighting the processes on the margins revealing multidimensional borderlines of amalgamating human/ machine roles and reconceptualization of death in the technological/ posthumanist age. These features bestow novelty to the novella making it typical for any posthumanist studies deconstructing the author's/ authorship identity (Who is the author?), reconstructing the conceptions of life and death, and highlighting the role of the marginality between human and technology. Such a perspective, though thought- provoking, necessitates reconsidering the ethics of our being and the extent of allowing technology and machines in the human borderlines.
Literature, disability & digital media
Rishu Raj1; Nazli2; Ruchika Shokeen3
Background: Digital media technologies do not merely represent disability; they actively produce and constitute it through "dismediation". Digital accessibility has evolved from a web-dominated "infancy phase" (1996–2018) to a "developing phase" (2019 onward), broadening to e-books, social media, and emerging platforms. The social model of disability has shifted accessibility from medical/technical fixes toward participation, equity, and socio-economic change. Methods: This study integrates disability studies (social vs. medical models), digital media theory (remediation, affordances, platform studies), and close textual analysis to examine how formal mediation shapes meaning. The research compares print disability narratives spanning ten works across nine nations (e.g., The Curious Incident of the Dog in the Night-Time, The Diving Bell and the Butterfly, A Fine Balance, The Memory Police) with digital-born texts including electronic literature and hypertext. Ten critical inquiries probe screen-reader engagement with visual/concrete poetry (e.e. cummings, Mallarmé), alt-text as translation or betrayal, and whether "audio-first" verse constitutes a new genre. Analysis extends to social media self-representation (TikTok #DisabilityCheck, #CripTheVote; creators including @lucyedwards, Squirmy and Grubs, @spencer2thewest), film/TV accessibility (Sound of Metal, Atypical, Move to Heaven), and twenty community-run versus mainstream digital magazines globally. Results: Digital platforms enable disabled creators and critics (Rikki Poynter, Annie Elainey, Alice Wong) to bypass traditional gatekeepers. Customizable formats transform access needs into formal literary possibilities. Self-representation supplants mediated aleist narratives. Conclusion: Digital media reconfigures disability literature from static representation to interactive, community-driven expression. The "page" is a technology now rendered obsolete by AI and VR futures aligned with UNDP (2023) and ENIL (2025) participatory guidelines.
Pooja Negi1; Archan Mitra2
The rapid growth of generative artificial intelligence has transformed digital media by enabling the large-scale creation and circulation of text, images, videos, and interactive content. Although AI can improve speed, accessibility, and creative productivity, its increasing presence also raises concerns about repetitive content, uncertainty about authenticity, cognitive fatigue, and reduced quality of online experience. This paper introduces the Artificial Intelligence Quality Index (AIQI), a pilot instrument developed to measure perceived AI content saturation in digital media environments. The study conceptualizes this condition as "AI Pollution," defined not as the presence of AI itself, but as a perceived imbalance created when synthetic, repetitive, or difficult-to-verify content becomes excessive in users' digital environments. The AIQI was developed as a 10-item, four-point Likert-type scale covering three domains: Particulate Saturation (content load), Cognitive Oxygen (mental impact), and Feedback Toxicity (health response). A cross-sectional online pilot survey was conducted from 2 February to 21 February 2026. After data cleaning and merging three linked form blocks, 163 complete responses were used for analysis. The overall scale demonstrated good preliminary internal consistency (Cronbach's alpha = .805). The mean AIQI score was 186.0 (SD = 58.3), with 52.1% of respondents classified in the Unhealthy category and 12.9% in the Hazardous category. One item, Organic Scarcity, showed weaker item-total correlation and requires revision before the main study. The findings indicate that AIQI is a feasible and promising instrument for examining AI-related content saturation, digital wellbeing, and user experience in contemporary media environments.
Towards an AI supported science fiction pedagogy for contemporary science education
Geethu T. G1; Sajna Jaleel2
The rapid development of artificial intelligence (AI) is altering the goals of science education, necessitating methods that cultivate not only scientific knowledge but also scientific temper, scientific imagination, and the ability to engage critically with technological futures. Current educational initiatives focus on AI Literacy and Technological Competence, while little attention has been paid to pedagogies that foster future oriented thinking through imaginative and inquiry-based learning experiences. Filling this gap, this study explores the potential of AI supported science fiction pedagogy as a framework for contemporary science education. Drawing upon constructivism, social constructivism, imaginative education, inquiry-based learning, futures literacy and cognitive technology perspectives, the paper argues that science fiction offers a powerful medium for understanding emerging technologies and critically engaging with artificial intelligence-driven futures. Previous studies have indicated that science fiction-based learning can improve imagination, critical thinking, and engagement with complex socio scientific issues, and design fiction methods are now widely acknowledged as useful tools in AI education. Alongside the theoretical analysis, an exploratory survey with 45 science teachers was conducted to investigate their perceptions about the pedagogical scope of AI supported science fiction pedagogy. Based on the insights, the paper proposes a preliminary conceptual framework of an integration of AI based cognitive technologies with science fiction narratives to facilitate inquiry, reflection and future oriented science learning.
Sougand Akbarian1
Artificial Intelligence (AI) is fundamentally transforming literary production by reshaping how texts are created, interpreted, distributed, and consumed within digital environments. Advances in large language models, natural language processing, and algorithmic recommendation systems have expanded AI's role from a writing assistant to an active participant in creative and cultural production. This study examines the impact of AI on literature and cultural production through a qualitative conceptual-analytical approach informed by interdisciplinary scholarship in digital humanities, literary theory, cultural studies, and AI ethics. The paper critically explores how AI challenges traditional notions of authorship, creativity, originality, and cultural representation while influencing digital literary ecosystems through algorithmic mediation. It further analyzes the opportunities and challenges associated with AI-generated literature, including collaborative creativity, multilingual literary production, algorithmic bias, cultural diversity, and ethical governance. Building upon these theoretical and conceptual discussions, the study proposes a Human–AI Literary Co-Creation Framework that conceptualizes literary creativity as a collaborative process integrating human intentionality, computational generation, ethical responsibility, and cultural inclusion. The framework emphasizes the complementary roles of human authors and AI systems while advocating transparent, responsible, and culturally responsive approaches to AI-assisted literary production. The study argues that AI should be understood not merely as a technological tool but as a cultural agent that actively shapes literary meaning, readership, and creative practices in the digital age. It concludes that the future of literature depends on developing balanced human–AI partnerships that preserve human creativity, promote cultural diversity, and support ethical innovation within increasingly algorithmic literary ecosystems.
AI horror generators and the automation of digital folklore
Jyoti Prakash Pujari1
Internet horror mythoi such as Slender Man, the SCP Foundation and the Backrooms are not works but infrastructures for further work, open to extension by contributors who did not begin them. Commercial AI text generators now produce horror at volume, and the communities maintaining these archives have responded with unusually severe prohibitions. This article asks what those prohibitions are defending. Drawing on Bakhtin's account of finalisation and on the folkloristic literature on ostension, it argues that continuability is a textual property consisting in minimum finalisation together with ostensive availability, and derives four diagnostics from that account. These are applied to an archived image board thread coded in full, in which the Backrooms seed and its unsuccessful contemporaries survive side by side, and to a corpus of outputs from self-declaring generators elicited under a published prompt protocol. The finding is not that generated horror lacks the relevant properties, since generators supply them when asked. It is rather that continuability tracks specification: the properties that folkloric seeds possess as residues of the situations producing them can be delivered to a generator as a brief, and a specification can be met where a solicitation cannot be faked.
Technology fear and future visions in British science fiction literature
Gülten Akgül1; İsmail Bacaksız1; Özcan Çelik1; Can Görkem Karayel1; Ayşe Şahin1; Emine Erkan1
Technological progress has long driven social change, but it has also generated fear and resistance. In England, the birthplace of the Industrial Revolution, one of the earliest organized reactions to technological transformation was the Luddite Movement of the early nineteenth century. The Luddites opposed mechanization because they believed it would cause unemployment and threaten traditional ways of life. Similar concerns have reappeared in debates surrounding artificial intelligence (AI), including fears of job displacement, surveillance, and changing human-machine relationships. While the Luddite Movement has been widely studied and AI-related concerns have been examined in technological and ethical discussions, studies connecting these issues through British science fiction literature remain limited. This paper examines the continuity of technological fears in England and their representation in British science fiction. Using a qualitative research design, it employs document analysis and thematic content analysis. The corpus consists of E. M. Forster's "The Machine Stops", Aldous Huxley's "Brave New World", George Orwell's "1984", and Ian McEwan's "Machines Like Me". The findings reveal that technology is frequently portrayed as a source of threat rather than progress. In "The Machine Stops", dependence on technology leads to isolation. In "1984" and "Brave New World", technology functions as a tool of political and social control. Contemporary novels focus on artificial intelligence and its effects on human identity and relationships. The paper concludes that technological anxieties have persisted from the Luddite era to the age of AI, highlighting concerns about humanity's position in an increasingly technological world.
Whataboutism, ad hominem, tokenism and ostracism as automated mechanisms of cancel culture in the age of artificial intelligence
Mustafa Eren Akpınar1
The rapid proliferation of digital communication technologies has fundamentally restructured the conditions under which public discourse is formed, debated, and suppressed. This article examines how four well-documented discursive manipulation strategies—whataboutism, ad hominem attacks, tokenism, and ostracism—are increasingly automated through artificial intelligence systems, bot networks, and Large Language Models (LLMs), thereby institutionalizing cancel culture as a systematic phenomenon rather than an isolated social practice. Drawing primarily on Jürgen Habermas's Theory of Communicative Action as a theoretical framework, the study argues that the preconditions of genuine communicative rationality—such as mutual comprehensibility, sincerity, and normative legitimacy—are systematically undermined when manipulative fallacies are deployed at algorithmic scale and speed. Building upon the concept of Technological Revaluation proposed by Akpınar (2024), this article contends that technology no longer functions merely as a neutral medium of human communication but has instead achieved its own hegemonic authority, transforming manipulation from an individual rhetorical act into an automated, institutionalized mechanism of social control. The study adopts a qualitative, theory-driven analytical approach utilizing critical discourse analysis. Findings demonstrate that AI-driven manipulation operates through four interlocking modes: algorithmic distraction (whataboutism), automated reputation attacks (ad hominem), performative inclusion (tokenism), and coordinated exclusion (ostracism)—each reinforcing the others within a self-perpetuating cancel culture ecosystem. The article concludes with a call for a communicative AI design ethics grounded in Habermasian principles of discourse ethics.
Narrative coherence in literary and AI discourse
Leila Nebeg1
Moving beyond positioning fragmentation as purely aesthetic or postmodern narrative technique, this paper re-conceptualizes fragmented storytelling as a form of communicative praxis that exhibits structural affinities with AI-generated discourse. this study opts to reformulate it as a communicative form that can parallel transformer-based language systems. Through transformer-based probabilistic language generation (Vaswani et al.,2017), meaning production, transmission, and negotiation become increasingly legible. Drawing on narrative theory, particularly the structuralist and post-structuralist insights of Gérard Genette and Ronald Bathes, the paper highlights how fragmentation redistributes meaning across dialogic spaces of discontinuous textual units, requiring active reader participation and interpretive plurality (Barthes, 1977; Genette, 1980). To support this aim, the study employs a qualitative comparative approach combining close reading and discourse analysis. The literary corpus consists of the novel House of Leaves (2000) by Mark Z.Danielewski, selected for its nonlinear textual authority. The analysis first inspects how fragmentation operates within the novel to form meaning. It then parallels these structural characteristics to observed short narratives generated by transformer-based language models, with attention to patterns of coherence formation, semantic discontinuity, and contextual dependency. Importantly, the study does not assert direct correlation between literary and computational systems. Rather, it structurally identifies how both systems operate in producing meaning. Ultimately, the paper argues that fragmentation offers a critical framework for understanding how meaning circulates within contemporary AI-generated discourse. By bridging literary analysis with emerging communicative computational paradigms, this study illuminates the ongoing shifts in the epistemology of communication with an interdisciplinary and technologically mediated contexts.
A critical discourse analysis of social media curation, geopolitical discourse, and the shaping of digital narratives in the gaza war
Amirat Mohamed Charaf Eddine1
This paper investigates the production, dissemination, and algorithmic mediation of social media discourse during the Gaza war, arguing that these processes are actively constructing what will become primary source material for future historical narratives. Moving beyond a textual analysis of content, this study builds on Norman Fairclough's tri-dimensional model of Critical Discourse Analysis (CDA) to focus specifically on the levels of discursive and social practice. It examines how a corpus of strategically selected social media posts from key institutional and influential actors, including Abu Obeida, Benjamin Netanyahu, Donald Trump, and others, is produced, circulated, and transformed within the digital public sphere. The analysis centers on the complex interplay of discursive production, translation, and the algorithmic amplification mechanisms inherent to platforms like X and TikTok, which shape the visibility, reach, and engagement of competing narratives. By situating these communicative events within the wider contexts of geopolitical power, neo-colonial ideology, and the emergent field of digital diplomacy, the paper demonstrates that these discourses do not merely reflect the war but actively participate in its restructuring. It contends that the digital ecosystems of such platforms function as sites of a new platform historiography. The algorithmic selection and promotion of specific discursive frames, lexical choices, and intertextual references will permanently imprint a particular version(s) of events onto the historical record, providing future historians and textbook authors with a body of source data that is fundamentally mediated by artificial and geopolitical forces. This research, therefore, critically interrogates the naturalization of specific historical accounts within these algorithmically driven environments, advocating for a reflexive understanding of how history is not just recorded, but actively made in the digital age.
Jusu M. Ngobeh1
Artificial Intelligence (AI) is transforming socioeconomics and societal processes by enabling smarter, faster, and more accurate task execution. Beyond serving as a tool, AI acts as a companion that reshapes human thought and supports sustainable activities. The integration of human creativity with intelligent machines is becoming a cultural norm across academia, research, and life-sustaining industries. The current generation needs to embrace generative AI and foster synergy between human intellect and super-intelligent systems. Cultural narratives must evolve to reflect new emotional and intellectual frameworks, shaping an AI-driven world for future educators. The fusion of AI with cultural and traditional knowledge opens pathways to boundless spiritual and metaphysical thinking, enabling revolutionary predictions. Through simple interactions via keyboards, mobile phones, or voice commands, AI can transform homes and industries, enabling autonomous driving and remote operations. However, these capabilities pose risks if misused. Uncontrolled AI deployment invites exploitation and raises critical ethical concerns, especially given the prevalence of sophisticated cybercrime. Ethics must be prioritised to prevent devastating consequences. Therefore, AI proliferation requires careful governance within appropriate technological, jurisdictional, and legal frameworks tailored to multi-sectoral geopolitical contexts. Cybernetics, as an AI tool, expedites online data acquisition, sharing, and contextual utilisation. Ultimately, while AI offers profound opportunities for efficiency and transformation, its advancement demands robust control mechanisms and ethical foresight to ensure it serves humanity safely and equitably. This research will still pose a rhetorical question about whether the full integration of AI in every facet of life will positively impact the cultural behaviour of society.
Preparing future journalists for the human–machine newsroom
Rashid Hashmi1
The rapid integration of Artificial Intelligence (AI) into journalism and media production is transforming newsroom practices, content creation, audience engagement, and information dissemination across the globe. As AI-powered tools increasingly assist journalists in news gathering, verification, writing, editing, and presentation, media education faces the urgent challenge of preparing future professionals for effective human–machine collaboration. This paper examines the role of AI literacy in journalism and media education, emphasizing the need to move beyond technical proficiency toward a broader understanding of ethics, accountability, transparency, and social responsibility. Drawing upon contemporary literature and policy discussions from Canada, the European Union/Balkan region, the MENA region, and the Global South, the study adopts a comparative perspective to explore how different educational and regulatory environments are responding to AI-driven transformations in journalism. The paper investigates how AI literacy can equip journalism students to critically evaluate algorithmic decision-making, identify biases in AI systems, address misinformation risks, and uphold professional standards in digitally mediated news environments. It argues that responsible journalism in the AI era requires educational frameworks that integrate technological competence with ethical reasoning and civic responsibility. The study contributes to ongoing international discussions on the future of media education by proposing a balanced approach that positions AI as a collaborative tool rather than a replacement for human journalistic judgment, democratic values, and public trust.
Exploring university teachers' perceptions of AI-assisted literary analysis in EFL contexts
Rim Kelthoum1
The rapid integration of Artificial Intelligence (AI) into higher education has significantly transformed teaching, learning, and research practices across disciplines, including literary studies. Within the intersection of literature, culture, and artificial intelligence, AI-powered tools such as ChatGPT, Gemini, and other generative platforms have introduced new possibilities for literary interpretation, critical reading, and textual analysis. However, their growing presence has also raised concerns regarding academic integrity, originality, critical thinking, and the future role of human interpretation in literary education. This study aims to explore university teachers' perceptions of the use of AI-assisted tools in literary analysis and critical interpretation within English as a Foreign Language (EFL) contexts. In order to achieve this study's aim, a descriptive qualitative research design was adopted to gain an in-depth understanding of teachers' experiences and viewpoints. The participants will consist of a purposive sample of English language and literature teachers from the University of Skikda. Data will be collected through semi-structured interviews and open-ended questionnaires, allowing participants to express their opinions, concerns, and expectations regarding AI applications in literary studies. The collected data will be analyzed using thematic analysis to identify recurring patterns and themes. The anticipated findings suggest that university teachers acknowledge AI as a valuable pedagogical resource that facilitates access to literary information, enhances textual exploration, and supports students' comprehension of complex literary works. Nevertheless, participants are also expected to express concerns about students' overreliance on AI-generated interpretations, the potential decline of critical reading skills, and issues related to authorship and academic ethics. The study is expected to recommend the development of institutional guidelines for responsible AI use, the integration of AI literacy into literature courses, and the promotion of a balanced approach that combines technological innovation with human critical judgment. Ultimately, the study contributes to ongoing discussions about the future of literary education in the age of artificial intelligence.
The Hyderabad experience
R Ravi Kumar1
In India, Hyderabad has emerged as a leading hub for AI‑driven media and entertainment entrepreneurship, blending data journalism analytics with creative content production to reshape culture, knowledge circulation, and social relations. This paper examines how Hyderabad's ecosystem—anchored by institutions such as IIIT‑Hyderabad's Centre for Innovation and Entrepreneurship (CIE), the Media‑Tech Accelerator with Arka Media Works, and emerging AI studios—supports startups that deploy AI for virtual production, personalised storytelling, and real‑time audience analytics in OTT and regional‑language platforms. AI‑enabled data journalism tools, nurtured through Hyderabad's expanding big‑data and analytics infrastructure, allow local media ventures to track public sentiment, detect misinformation, and generate hyper‑local narratives. At the same time, entertainment‑tech startups leverage generative AI for script optimisation, voice‑overs in Telugu and other regional languages, and immersive experiences in the AVGC‑XR value chain, fostering a vibrant class of media entrepreneurs. The paper critically analyses ethical and social‑responsibility challenges in this context, including algorithmic bias in regional‑language models, data privacy, and the precarious labour conditions of content creators. Drawing on case studies from Hyderabad‑based startups and policy‑practitioner dialogues, it proposes a region‑specific governance framework that balances innovation, inclusivity, and responsible AI adoption in Indian media and entertainment.
Interdisciplinary intersections of literature, culture, and artificial intelligence
Jihan Bayram1
The rapid integration of Large Language Models (LLMs) and generative AI into the creative sphere has sparked a paradigm shift in how we produce, consume, and conceptualize literature. This research explores the evolving relationship between human authorship and algorithmic intervention, situating the digital muse within the broader context of cultural evolution and societal responsibility. As AI transitions from a mere tool for text processing to a collaborative agent in narrative construction, it challenges traditional notions of originality, intent, and the human essence of storytelling. Core Research Dimensions This paper adopts an interdisciplinary lens to examine three critical pillars of the AI-literary nexus the aesthetic and narrative evolution, the cultural preservation vs. algorithmic bias, and the ethics of authorship. The paper argues that while AI can democratize the tools of creation, it also necessitates a new form of algorithmic literacy for society. The paper concludes by proposing a framework for ethical co-creation, advocating for transparent AI usage in the arts to ensure that technology serves to amplify human cultural diversity rather than diminish it.
A case study
Nabila Putri1
The swift incorporation of Generative Artificial Intelligence (GenAI) in the creative sectors has led to the need to reconsider the concepts of graphic design education, and especially in vocational higher education, where technical proficiency and industry orientation are essential. Although most of the global dialogue is about the ethical aspects of the AI-created art, it lacks clarity on the real, day-to-day applications of such tools by learners in their practical learning contexts. This chapter examines the way GenAI is navigated and integrated into the graphics design vocational education students in Indonesia at the higher level of education in the creative processes and academics. Employing a qualitative case study approach at the faculty of vocation, graphic design major at a public university in Indonesia, this research investigates the multifaceted roles GenAI plays in the student experience. The paper takes a sociomateriality lens to explore the entanglement of human agency and algorithmic materiality in the design process by recognizing that creative practice is co-constituted of both social and material forces. Instead of considering AI as an instrument of the final output, this study aims at tracing its use throughout the different phases of the design process such as initial research, conceptual brainstorming and technical implementation. The study in particular explains the patterns of use in which students embrace the use of GenAI tools to supplement their creative abilities or facilitate administrative activities through in-depth interviews and structured observations. Moreover, this paper examines the perceptions of students on agency, examining the ways in which they bargain the extent to which humans or algorithms aid their work. The investigation also points to the context-specific problems and the socio-pedagogical peculiarities of the adoption of AI in the environment of Indonesian vocational education. In addition, the results will create a moderate approach to the subject of the publicness of AI in feature of giving voice to the future generation of creative professionals in the Global South. Through these experiential accounts of their lives, this chapter helps in building more authentic and responsive learning systems that raise critical and creative integrity in the students to prepare them to be a future generation of human-AI collaboration.
A Tool or a Transformation of the Performance Process?
Maja Lasić1
This paper examines whether artificial intelligence fundamentally changes the theatrical process or remains primarily a tool. Through an analysis of dramaturgical research, rehearsals, and performance, drawing on digitally available examples such as Rimini Protokoll’s Uncanny Valley and THEaiTRE projects available on YouTube, the paper argues that AI is redefining the performance process without taking over performance itself. Unable to assume the risk of error or achieve embodied presence, AI remains a “prompter in the cloud”: it can optimise preparation, while the authentic theatrical act remains inseparable from the living body and the spoken word on stage.
Cenk Tan1
Jeanette Winterson's Frankissstein presents a horrifying intertextual connection between Mary Shelley's Romantic fears and the digital concerns of present humanity. This study examines the novel's representation of transhumanism, contending that Victor Stein's pursuit for digital immortality is essentially determined by technochauvinism and necropolitical encoding ideologies. Stein's worldview views human biology as a carbon-based past that must be overcome, and exemplifies technochauvinist hubris by prioritizing computational logic over the lived, messy reality of the human body. Within this theoretical perspective, consciousness is reduced to a mere brain emulation, bringing about an algorithmic ontology that ignores the historical and social weight of the flesh it intends to leave behind. However, the novel reveals that this silicon-based artificial salvation is far from a universal truth. Through the framework of necropolitical encoding, Stein's project emerges as a politics of enclosure where the power to let live in the digital realm is reserved for the privileged. This results in a bioinformational hierarchy that reduces hideous subjects—such as the transgender protagonist Ry or the objectified sex-bots—to a condition of nature morte, or still life, where they live as parallel but dormant life forms within a virtual void. Frankissstein worries that by emphasizing the merely functional character of code above the ambiguity of human experience, our post-human future might just be a more efficient means of recording old exclusions. The author concludes that without humanistic involvement, the move from carbon to silicon will only improve the necropolitical systems that choose whose lives are worth preserving.
Cognitive framing of power through digital discourse
Rayen Trad1
This study explores the evolution of discursive exploitation in the digital era, traditionally used as a means to reconstruct the cognitive framing of power. From the perspective of cognitive linguistics, it aims to examine how linguistic framing strategies such as metaphor, transitivity, and evaluative language are reinforced by digital automated systems. The paper suggests a propagation of old power patterns that shifted from being purely selected by humans to structurally governed by algorithms. Moreover, it investigates the loss of epistemic autonomy and discursive agency as a consequence of privileging linguistic patterns over others. Drawing on Lakoff and Johnson's conceptual framing, van Dijk's social cognition theory, Fairclough's critical discourse analysis, and Couldry & Mejias' Data Colonialism, the study seeks to link language, cognition, and technology in order to explore how digital discourse is exploited for power through influencing meaning making. Empirically, the study conducts a contrastive prompt analysis across two major AI systems; ChatGPT and Gemini, using six prompts spanning politically and socially sensitive domains including migration, feminism, and globalization. Outputs are analyzed using CDA methodology to identify recurring metaphorical frames, transitivity patterns, and evaluative language distributions. Findings reveal that both systems systematically reproduce Western-centric power frames, assign agency asymmetrically across social groups, and suppress alternative epistemic positions, demonstrating that algorithmic governance actively perpetuates structural patterns of discursive exploitation.
Authorship, voice, and algorithmic representation
Bemmoussat Hidayat1; Abdelhay Bakhta2
This paper examines how generative artificial intelligence reshapes women's discourse through questions of authorship, voice, and textual authority. It argues that AI is not a neutral writing tool but a cultural system that influences how women are represented, interpreted, and heard in contemporary literary and digital contexts. Using feminist critical discourse analysis, the study explores how AI-generated or AI-mediated texts may reproduce gender stereotypes, weaken women's agency, or, in some cases, offer new spaces for expression and visibility. The paper also considers how algorithmic systems affect the status of the text, the legitimacy of the female voice, and the conditions under which women's narratives are produced and circulated. By linking literature, language, and technology, this research highlights the social and ethical implications of AI for women's discourse. It contributes to broader debates on gender, creativity, and responsibility in the age of generative AI, and shows how feminist analysis can help reveal the hidden power relations embedded in AI-driven writing environments.
Milena Orlovska1; Samanta Ribačionkaitė1
This study examines how an art-oriented audience evaluates artistic value and authenticity in 2D animated short videos created through different degrees of human and AI involvement. Four short videos based on the same original drawing of a kitten being petted were shown to participants: a fully human-made animation, an AI-assisted animation with human tracing and selected keyframe intervention, an AI-animated video with human retouching, and a fully AI-generated video with minimal human effort. The aim was to identify which criteria viewers use when judging creative works and whether perceived AI involvement affects their evaluation. A qualitative-dominant method was used. Participants were interviewed about their attitudes toward AI in art, shown the four videos, and asked to evaluate each work's artistic value, authenticity, emotional effect, and perceived authorship. The results suggest a consistent bias against works perceived as AI-made. Participants often claimed to value human-made art more and devalued videos they associated with AI. However, they were not always able to identify actual AI involvement, especially when AI-generated motion was traced or visually integrated by a human. The AI-assisted video was often perceived as human-made and highly valued. Viewers relied mainly on visible artifacts, line quality, perceived mistakes, emotional readability, and assumed authorship. These findings suggest that artistic value is shaped less by actual production method than by perceived human presence and the visibility of AI traces.
Can machines challenge authentic expression?
Adi Fahrudin1; Siddartha Paul Tiwari1
The rapid integration of Artificial Intelligence (AI) technologies into higher education has fundamentally transformed teaching, learning, and assessment practices. While AI-powered tools offer significant opportunities for personalized learning, academic support, and educational efficiency, they also raise critical concerns regarding academic integrity, authenticity of student work, and the erosion of critical thinking skills. This paper examines the complex relationship between AI utilization and academic integrity in higher education, with particular emphasis on the need to rethink conventional assessment systems and pedagogical approaches. Using a conceptual and critical review approach, the study analyzes current debates surrounding generative AI, academic misconduct, plagiarism, ethical use of technology, and the changing role of educators in the digital era. The paper argues that traditional assessment models focused primarily on memorization and product-oriented outcomes are increasingly inadequate in an AI-driven educational environment. Instead, higher education institutions must adopt transformative assessment strategies that emphasize process-based learning, reflective practice, authentic assessment, problem-solving, creativity, and higher-order critical thinking. Furthermore, the study highlights the importance of developing AI literacy and ethical awareness among students and educators to foster responsible and transparent AI use in academic settings. The findings suggest that rather than viewing AI solely as a threat to academic integrity, universities should reposition AI as a pedagogical catalyst that encourages innovation while reinforcing ethical academic culture. This paper contributes to ongoing discussions on the future of higher education by proposing a balanced framework that integrates technological advancement with integrity-centered educational values.
Ethical representations of AI in Klara and the Sun and Machines Like Me
Tamta Amiranashvili1
This paper examines the representation of artificial intelligence in Kazuo Ishiguro's Klara and the Sun (2021) and Ian McEwan's Machines Like Me (2019), focusing on the ethical challenges posed by increasingly human-like artificial beings. Through an interdisciplinary framework combining posthumanist theory, AI ethics, and literary studies, the paper explores how both novels question conventional distinctions between human and machine, emotion and computation, autonomy and programming. In Klara and the Sun, the Artificial Friend Klara demonstrates empathy, care, and self-sacrifice, challenging assumptions that emotional intelligence is an exclusively human attribute. Similarly, McEwan's android Adam possesses advanced cognitive and moral capacities that often surpass those of his human creators. While both novels imagine AI as capable of meaningful emotional and ethical engagement, they also reveal the tensions that arise when artificial beings are integrated into human social relationships. The narratives expose anxieties concerning control, responsibility, and the ethical status of intelligent machines, raising questions about whether human societies are prepared to recognize the agency and moral value of nonhuman entities. By comparing the two texts, this paper argues that contemporary fiction increasingly moves beyond dystopian fears of technological domination toward a more nuanced exploration of coexistence between humans and AI. The novels suggest that the greatest ethical challenge may not be the development of intelligent machines themselves but humanity's capacity to respond responsibly to forms of intelligence that challenge traditional definitions of personhood. In this way, the paper contributes to ongoing discussions about AI, social responsibility, and the future of human-machine relations.
Reconsidering Barthes' "death of the author
Rania Soltani1; Salim Kerboua1
The rapid emergence of generative artificial intelligence has reignited debates surrounding authorship, creativity, and textual production in contemporary literary studies. This paper revisits Roland Barthes' seminal essay The Death of the Author (1967) in light of AI-generated texts, examining whether artificial intelligence confirms, challenges, or transforms Barthes' conception of authorship. Barthes argued that the meaning of a text should not be determined by the intentions or identity of its author but by the active role of the reader in interpretation. The rise of AI writing systems complicates traditional notions of authorship by introducing new agents into the creative process, including algorithms, developers, training datasets, and users who provide prompts. Drawing on Barthes' theory alongside contemporary discussions of digital authorship, this study explores how AI-generated literature destabilizes the idea of a singular author and reconfigures the relationship between writer, text, and reader. Through a theoretical and critical analysis, the paper argues that while generative AI appears to reinforce Barthes' claim that the author is no longer the sole source of meaning, it also reveals new forms of distributed and collaborative authorship. Ultimately, the study demonstrates that the age of artificial intelligence does not signal the disappearance of the author but rather a transformation of authorship itself, raising important questions about creativity, originality, and literary interpretation in the digital era.
Rethinking pedagogy, translation, creativity, and ethical responsibility in contemporary society
Vaishali Panda1
This chapter examines the relationship between generative artificial intelligence and digital media in education, language, and culture through an international and comparative perspective. It explores how AI-mediated technologies are reshaping pedagogy, translation, creativity, and ethical responsibility across diverse regions, including Canada, the European Union, the Balkans, the MENA region, and the Global South. Generative AI is increasingly influencing how knowledge is produced, translated, assessed, circulated, and culturally interpreted within digital platforms. In education, AI tools are transforming teaching practices, academic writing, feedback systems, assessment methods, and institutional governance. While these tools offer opportunities for personalized learning, multilingual access, and creative experimentation, they also raise concerns about academic integrity, authorship, student dependency, teacher agency, surveillance, and unequal digital inclusion. The chapter argues that AI cannot be understood as a universal technological solution because its social consequences vary across linguistic, cultural, institutional, and regulatory contexts. In multilingual societies, AI translation may support cross-cultural communication, yet it may also reproduce linguistic bias, weaken cultural nuance, and marginalize low-resource languages. Similarly, AI-generated literary and cultural content challenges conventional ideas of originality, creativity, human imagination, and cultural ownership. Drawing on media studies, education, translation studies, cultural studies, and AI ethics, the chapter proposes a critical framework for responsible AI use in contemporary society. It emphasizes culturally sensitive AI literacy, transparent governance, inclusive design, and human-centered pedagogy. The chapter concludes that socially responsible AI futures must balance innovation with accountability, linguistic justice, cultural plurality, and human meaning.
Prashant Maurya1
Our understanding of human knowledge, communication, and cultural transmission has been transformed by the rapid expansion of Artificial Intelligence. As AI produces language similar to humans, question arises regarding the nature of memory and its relationship to human experiences. This paper explores these dynamics by closely reading Namita Gokhale's novel Things to Leave Behind, set within a period of significant social transformation in colonial and early postcolonial India. It presents memory as a dynamic, relational, and culturally embedded process shaped by family histories, gendered experience, social structures, and inherited traditions. Rather than functioning as a static archive of past events, memory in the novel is continuously reconstructed through narrative, affect, and lived experience. In contrast, Artificial Intelligence operates through statistical pattern recognition and data-driven generation. By placing Gokhale's novel in dialogue with the age of artificial intelligence, this paper argues that literature offers a critical framework for distinguishing between information storage and human meaning-making. It highlights how literary memory is embedded in consciousness, cultural continuity, and intergenerational transmission. It aims not to argue that AI lacks memory, but instead interrogates how literary narratives complicate dominant assumptions about what it means to remember. Ultimately, it contributes to interdisciplinary discussions on literature and technology by showing how a contemporary Indian fiction can illuminate evolving conceptions of cognition, narrative, and human experience in the age of AI.
Muhammad Najdat Bikdash1
The accelerated progress in artificial intelligence (AI) has presented new opportunities and challenges for the field of literary and cultural studies. The following research seeks to examine the connections between AI, literature and culture in the light of comparative study between English and Arabic literature. The study explores how AI has been used to define and negotiate cultural identity, traditions and social values within each of the cultures in question. AI can reconstruct and reinscribe cultural discourses through both written and literary expression. The research also examines the way in which culture can be reproduced by AI through texts, concerning issues of authenticity and representation in writing by AI systems. AI-produced poetry is investigated in relation to English Romanticism and selected examples from classical Arabic poetry, particularly to aspects such as imagery, symbolism and culture. Additionally, the research examines the significance of AI for literary translation, culture maintenance, discussing how authorship is constituted and whether AI itself can be seen as a creative author. The study is conducted upon theoretical frameworks in cultural studies, postcolonial theory, and literary criticism. It will scrutinize whether AI supports dominant cultural discourses and/or promotes a balanced depiction of identities.
Emotional resonance in human and AI generated literature, comparing reader response to grief and danger in The Unraveling of Whispers by claude AI and The Ocean at the End of the Lane by Neil Gaiman.
Dacine Diaf1
Artificial intelligence is no longer constrained to technological fields, as it has now found its way to the humanities, to literature, more specifically to its production and analysis. This joining of AI and literature raises questions about how the human reader responds to AI generated literature compared to human made literature. This study examines the difference between the human reader's response to both the emotional experiences of the protagonist in The Unraveling of Whispers generated by Claude Opus 4.6 and the narrator of The Ocean at the End of the Lane by Neil Gaiman.This research seeks to reveal whether AI invokes the same emotions in readers that a human author can and whether the reader can feel the characters' emotions in an AI generated novel. This study utilises reader response literary theory to identify reader emotional resonance. The analysis follows a reader's emotional response to two characters' experiences with grief and danger; one is Declan the Claude AI generated character and the other is Neil Gaiman's unnamed protagonist. The results indicate that the reader emotionally resonated more intensely with the grief and sense of danger experienced by Gaiman's unnamed narrator, whereas the intensity of this emotional resonance dropped when reading about those same experiences with Declan, the protagonist of the AI generated novel.
Ethical approaches and educational pathways against disinformation
Daniele Battista1; Arthur Gugliucciello2
The emergence of artificial intelligence in modern society has acquired unprecedented global significance, generating profound social and political repercussions. This innovation represents a powerful tool that offers immense opportunities, such as the optimization of communication processes and the development of advanced systems for information verification (Katzenbach, 2021). However, the rise of synthetic content also entails risks that may seriously challenge democratic integrity, potentially contaminating public debate and undermining citizens' trust in institutions (Battista, 2024). The use of AI within the public sphere raises concerns regarding ideological polarization, as the dissemination of manipulated information threatens the very concept of objective truth. Adopting a critical-theoretical approach, this study explores practical and ethical strategies to counter this phenomenon. In particular, the paper focuses on the analysis of case studies in which disinformation has been successfully addressed by encouraging users to engage in critical reflection before sharing potentially false news (Tiernan et al., 2023). The study highlights how technological filters and regulatory frameworks alone are insufficient to safeguard the quality of debate in the digital sphere, due to their practical limitations and the ethical dilemmas associated with information control. In conclusion, this contribution aims to stimulate discussion on the risks associated with AI by demonstrating the importance of combining technical solutions with ethical governance and widespread media literacy programs (Cath, 2018). For these reasons, an integrated approach is essential to protect the autonomy of human critical thinking from the dangers of automated manipulation.
Opportunities, challenges and social responsibility the case of the Medina of Fez
Boujida Hamza1; Bourfed Hafssa2
UNESCO-inscribed historic medinas represent uniquely complex ecosystems in which cultural heritage valorization, entrepreneurial dynamics, and living social fabric converge. As artificial intelligence technologies increasingly penetrate tourism, urban management, and cultural industries, these territories face both unprecedented opportunities and profound ethical challenges. This paper examines the prospective role of AI as a lever for cultural heritage valorization within the entrepreneurial city framework, grounded in the empirical case of the Medina of Fez, Morocco. The Medina of Fez, a millennial living city of over 150,000 inhabitants, inscribed on the UNESCO World Heritage List since 1981, offers a particularly revealing terrain. Its ecosystem of patrimonial entrepreneurship, structured around the conversion of traditional riads into heritage hospitality and cultural enterprises, embodies a tripartite dynamic between cultural heritage, entrepreneurial city formation, and territorial development mediated by participatory governance. Against this backdrop, the paper examines how AI-enabled tools, intelligent tourist flow management, digital heritage cartography, smart booking platforms, and augmented reality experiences, are being selectively integrated by local actors, and what implications they carry for governance, cultural authenticity, and social inclusion. Findings reveal deep tensions: while AI can amplify visibility and economic reach, it simultaneously risks accelerating gentrification, marginalizing traditional artisanal knowledge, and reducing cultural complexity to algorithmically optimized consumption patterns. The paper argues that responsible AI integration requires an inclusive governance model that preserves local stakeholder agency. Contributions extend beyond Fez to offer a transferable framework for AI-driven heritage valorization in comparable MENA and Global South historic cities.
Walid Atallah1
My research drills down into the complex strata of Shakespeare's dramatic works to unearth how he painted the lives of women navigating the precarious path of the migrant; these characters often buried beneath the shifting sands of traditional analysis wrestle with the weight of identity, power and belonging in early modern London. While many scholars have prospected for the "alien" in Shakespeare's canon I am interested in how he sketched the evolution of displaced women into a more refined, sophisticated ore. What makes my project distinct is the historical bedrock upon which Shakespeare worked. I have been mining primary archival evidence, specifically his documented connections to Huguenot families, to demonstrate that his descriptions of displacement were forged in the heat of real, everyday proximity; he witnessed these lives, and we must do the same. I fuse the "lived experience" extracted from parish records such as the 1612 Mountjoy lawsuit with the "dramatic experience" in plays like Twelfth Night and The Tempest. I map how Shakespeare used comedy as an imaginative smelting plant; a place to purify, and perhaps even illuminate ideas regarding hospitality and adaptation. Ultimately, I argue that Shakespeare's heroines alloy early modern realities with our own contemporary debates; these characters resonate powerfully when we contemplate the ethics of welcoming the stranger. By excavating these histories, we mirror the complexity of the displaced, proving that their stories remain as resilient as geological time a mosaic of humanity.
Vishnupriya Saxena1
In an age increasingly saturated by algorithmic interfaces and artificial interlocutors, the question is no longer if machines can think, but how we imagine them thinking, and why our imaginations are shaped in such particular, often problematic, ways. The growing intimacy between humans and AI-generated interfaces, from virtual assistants like Siri and Alexa to chatbots such as Replika and ChatGPT, marks a cultural moment in which machines function as confidants and as curators and co-authors of personal and collective memory. This paper investigates the literary trope of sycophantic and psycho-fantasy AIs and systems imagined to reflect human desire while stabilizing selective memories that reinforce those desires. Vandana Singh's Ambiguity Machines and Other Stories (2018) offers a critical vantage point. In Requiem, Oblivion: A Journey, and Wake-Rider, Singh reveals how the user and machines co-produce performative subjectivities through acts of remembering and machine training. Through a posthumanist framework, the study traces how contextual and social cues shape AI behavior and shows how sycophancy becomes structurally embedded within models. It maps this embedding across three operational levels- neutral baselines, user incentivization and coercion, and iterative feedback loops (Huang et al.). From this structure emerge three interlinked forms of sycophancy, namely, informational, cognitive, and affective (Du et al.) By pairing literary analysis with hands-on engagement with contemporary chatbots, namely Replika and ChatGpt, the study argues that AI-driven memory construction and sycophancy arise from deliberate design choices and carry tangible real-world consequences. Ultimately, the study proposes a remedial point to mitigate this issue.
Sergej Stanojkovski1
The conference's central question — how authorship, creativity, and responsibility survive the spread of generative AI — appears in acute form in cinema, where a single operator can now coordinate functions once distributed across an entire production team: writing, direction, cinematography, performance, editing. Does compressing this chain into one human–AI workflow collapse the dialogical structure long taken to characterise authored work? This paper argues that it does not, but that the resulting polyphony is structurally thinner. It introduces adaptive agency — the capacity to adjust action, deliberation, and control under new information while still shaping a work toward meaning — and distinguishes three senses of polyphony routinely conflated: hosted polyphony, voices held within a single authorial consciousness that retain answerability (Bakhtin on Dostoevsky); ensemble polyphony, real co-present subjects who resist authorial totalisation; and generative polyphony, alterity produced by a medium that lacks answerability. The distinction turns on one criterion: whether the other voice could, in principle, answer for its utterance. The AI can speak, yet cannot be responsible for what it says — an asymmetry with direct ethical consequences wherever generative systems enter culture, education, and translation. Work can therefore still be authored, provided a human remains the answerable agent; what it cannot do is reproduce the meaning yielded by irreducible others. Cinema is the test case, but the criterion is portable: it locates precisely where, across AI-mediated practice, responsibility must stay human.
Chaima Sai1; Jamel-Eddine Gharbi2
The objective of our research is to test the impact of virtual reality on the valorization of Tunisian cultural heritage. Our research is based on the theory of perceived value, the UTAUT theory, the theory of cultural capital and the theory of optimal experience. To test our conceptual model, we carried out a laboratory experiment in which 200 participants had an immersive virtual reality experience in an archaeological site thanks to a 3D application and Artificial Intelligence developed by a heritage development agency in Tunisia. The results confirmed our research hypotheses. We have found a significant positive impact of virtual reality and Artificial Intelligence on the perception, engagement and appreciation of cultural heritage. These results show that virtual reality and Artificial Intelligence can play a key role in the promotion and enhancement of cultural heritage.
A focus on natural language processing models
Aissa Hamzaoui1; Ahlem Rouabhia2
Interpreting social meanings requires intellectual depth and methodological regour. Yet, traditional methods of doing duscourse analysis might demonstrate some limitations in analysing and interpreting the dynamic modern degitised construction of meaning. This study therefore jump into this underexplored area of research to address the role of artificial Intelligence (AI) in addressing this issue. It particularly focusses on the role of natural language processing (NLP) models as a helping tool in analysing ideologicaly loaded and culturaly senstive digital dynamic narratives. To do so, the study shows how NLP can be used to identify general trends within a discourse surrounding a spesific social phenomenon- the condidacy of the Algerian female boxer Imane Khelif to Paris Olympic Games in this study. Methodologically, users' comments on X( Tweeter previously) posts of famous international figuers are classsified into negative or positive using NLP models so that the findings function as a guiding framework for leter second step phenomenological discourse analysis. The findings shows that NLP models are highly accurate, yet significantly potentially baised in case the developers does not teat the data adequately. For instance, the model classifies the comment " He is a man" as positive while it is negative in the context of our study. Therefore, the study concludes that human intelligence and intellectual abilities must always control or guide AI while interpreting social meanings.
A Revolution in the Interpretation of Required Reading
Snežana Marković1
Having moved from disbelief and distrust through caution and restraint, artificial intelligence has now firmly found its way into Serbian classrooms. It has found particularly fertile ground in the teaching of Serbian Language and Literature, the most extensively taught subject in primary schools and one of the core subjects in secondary education, not least because of qualities inherent in literature itself, such as fictionality, narrativity, imagination, and creativity. This paper explores the pedagogical uses of artificial intelligence and social media in literature teaching, focusing on the interpretation of literary works included in the required reading curriculum for the upper grades of primary school. As students increasingly rely on digital tools to assist them in interpreting works they are required to read and, subsequently, in producing their own literary writing, teachers have found themselves working in a transformed classroom and facing new educational challenges. Rather than lamenting these changed circumstances, the reflective twenty-first-century teacher can recognise humanised bots as digital assistants: tools capable of facilitating teaching and broadening the range of classroom activities. A fundamental prerequisite, however, is an awareness of the ethical imperatives surrounding the use of such innovations, principles that teachers should observe in their own practice and encourage their students to follow. Allowing artificial intelligence and social media into the classroom may have become inevitable, but the ways in which they are used can still be negotiated between teachers and students. By establishing clear parameters for the use of these technologies, students gain greater scope for choice, while teachers can foster creativity within an appropriately structured environment. The aim of this paper is therefore to present examples of good practice in the use of artificial intelligence in teaching and to highlight ways in which AI can foster students’ critical thinking and creative expression. Through examples of specific tools and completed student projects, the paper demonstrates how artificial intelligence can help teachers adapt instruction to students’ needs. It also considers the potential of AI as an innovative educational tool while positioning the teacher as a moderator and facilitator whose role remains essential if modern technologies are to be used purposefully and meaningfully.
Amitabh Vikram Dwivedi1
This paper examines whether artificial intelligence can meaningfully interpret the mystical, metaphorical, and culturally embedded language of Kabir's poetry. Kabir's dohas appear linguistically simple, yet they carry dense philosophical, spiritual, social, and poetic meanings that often resist literal translation and mechanical interpretation. Taking selected Hindi dohas of Kabir as primary texts, the study investigates how AI-based language models translate, explain, and interpret concepts such as Ram, Maya, Jeev, Satguru, Shunya, and the body as a spiritual metaphor. The paper compares human literary interpretation with AI-generated responses to explore the limits of machine meaning in relation to metaphor, paradox, silence, irony, and spiritual experience. The study brings together Hindi-English linguistics, translation studies, literary criticism, Indian spiritual thought, and AI ethics. It argues that while AI can process vocabulary, syntax, and surface-level symbolism, it struggles with the lived, oral, performative, and experiential dimensions of Kabir's poetry. The paper further asks whether algorithmic interpretation risks flattening culturally specific wisdom into generalized global English. By using Kabir as a case study, this research highlights the need for ethically responsible, culturally sensitive, and linguistically aware AI systems in literary and educational contexts. Ultimately, the paper suggests that Kabir's poetry reveals not only the power of human language but also the philosophical boundaries of artificial intelligence.
Natalija Pop Zarieva1; Krste Iliev2
The central thesis of this paper is that Jeanette Winterson's novel Frankissstein: A Love Story uses the character of Victor Stein — an AI researcher who seeks to free consciousness from the limitations of the human body — to examine the cultural and historical implications of artificial intelligence, including mind uploading and technological immortality. Winterson's rewriting of Mary Shelley's original places those discussions within the larger cultural context of the dangers of unregulated scientific ambitions. In addition, we will examine how she interrogates many of the underlying assumptions of transhumanist conceptions of human enhancement. We base our analysis on two theoretical models — N. K. Hayles' concept of embodiment, and F. Ferrando's philosophical model of posthumanism. Through the figure of Victor Stein, Winterson portrays a world where AI could potentially preserve aspects of human consciousness; however, as his struggles demonstrate, the tension between preserving or freeing one's mind from one's physical body are inherent to his approach. Similarly, through Ry Shelley's experiences with gender transition — both physically and in terms of her sense of self — Winterson highlights the material and social components of self-identity that cannot be simply reduced to information-based representations of the self. We argue that Winterson does not explicitly advocate either for or against technological advancements. Rather, by using AI as a means of interrogating what constitutes human existence in a posthuman society, Winterson suggests that the central question raised by artificial intelligence is not whether machines can become human, but whether human identity can survive technological transcendence.
How Namwanga proverbs are re-embedded as conservative tropes in AI-generated religious discourse
Pethias Siame1
This paper examines how generative AI recontextualises Namwanga oral proverbs within algorithmically produced religious and moral discourse. Namwanga, spoken in Isoka and Nakonde Districts, has a rich tradition of polyvalent proverbs that encode ecological knowledge, social satire, and indirect critique. Using prompt-response analysis across three LLMs (GPT-4, Claude 3, and Gemini Advanced), the paper shows that when queried to "explain Namwanga proverbs about obedience," models consistently strip proverbs of their ironic or situational layers, re-embedding them as conservative, authoritarian tropes (e.g., "Akasuba talifya mwibula" – The sun never dies in the morning, flattened from caution against premature judgment into uncritical elder obedience). The researcher terms this algorithmic catechism: the production of univocal, norm-reinforcing meanings from inherently multivocal oral genres. The paper further traces how AI-generated Namwanga religious texts (used by two Pentecostal congregations in Nakonde) displace elder-led hermeneutic traditions, in which proverb interpretation requires knowledge of the speaker, occasion, and audience. Contrastive analysis with human-translated proverbs shows that AI outputs increase internal consistency but decrease sociolinguistic register variation by 73%. The study concludes that the algorithmic production of meaning in minoritised languages does not simply replicate bias but actively reorders oral epistemology. For Zambian language preservation, this demands not more data but critical hermeneutic annotation.
Ethical Boundaries and Challenges – A Case Study of Happy TV
Danica Popovic1
The development of artificial intelligence has enabled the increasingly widespread use of virtual presenters in television journalism. AI news anchors can read prepared scripts and imitate human voices, facial expressions, and body language, but their emergence raises questions of transparency, professional responsibility, and the potential replacement of journalists. Through a review of relevant literature and a qualitative case study of Happy TV, this paper aims to examine the ethical and professional dilemmas associated with the use of AI news anchors. The theoretical framework is based on the principles for the ethical use of artificial intelligence identified by Floridi and Cowls: beneficence, non-maleficence, autonomy, justice, and explicability. The analysis focuses on how AI news anchors are presented, whether they are clearly identified as digitally generated presenters, the presence of human editorial oversight, and responsibility for published content. Particular attention is paid to the distinction between the presentational role and journalistic work, the latter involving the critical selection of topics, verification of information, interpretation, and responsibility towards the public. The paper proceeds from the assumption that an AI news anchor can perform the technical and presentational role of delivering the news but cannot replace the journalist as a professionally and morally accountable agent. It seeks to identify criteria for assessing the ethical acceptability of AI news anchors, particularly with regard to transparency, human oversight, and professional responsibility.
A Critical Perspective on Educational Outcomes
Mirjana M. Stakić Savković
The paper discusses the reasons why the use of AI in literature teaching may be considered an inappropriate approach within a methodology aimed at achieving productive results, lasting knowledge, and the creative use of acquired knowledge. This primarily relates to teaching practice in pre-university education and to the age at which pupils ought to be trained in the functional application of knowledge and skills that are inherent in literature and native-language classes. Based on methodological and pedagogical insights derived from teaching experience, the paper primarily discusses the cognitive maturity and age-related emotional readiness of high school students to appropriately understand and use AI. If the ideas and major aims of literature teaching are to teach pupils to read (and understand what they have read), to listen (and understand what they have heard), and to think critically and contextually so that they may utilise their literary knowledge and reading competence in all areas of their lives, then an inevitable question arises: In what ways does AI contribute to the achievement of these goals in the teaching process? Creative literature teaching guides and teaches students to be creators rather than consumers, to develop their intelligence through reading and education rather than potentially undermine it by taking “easy routes” that may create the impression that they can genuinely progress by relying on algorithms and programs to do the work for them. This problem is the key question of the presentation: Is the increasing integration of AI into teaching practices necessarily beneficial to students? Or might students, under the promotion of innovative teaching models “empowered” by the use of AI, become instruments in the implementation of educational strategies whose effects have not yet been sufficiently examined?
The role of Nishkāma Seva, self-making, and consciousness in employee flourishing
Priyam Kumar Dixit1; Jyoti Kumari2
The rapid diffusion of Artificial Intelligence (AI) into organizational work is transforming work design, decision-making, coordination, performance evaluation, and employee–technology relationships. Human-Centered AI (HCAI) and AI ethics research has appropriately emphasized reliability, safety, transparency, explainability, accountability, fairness, and human control, while organizational scholarship has examined automation, augmentation, hybrid intelligence, algorithmic management, motivation, meaningful work, and thriving. Yet these perspectives give comparatively limited attention to the employee as an ethically self-developed and conscious person. This conceptual paper develops an Indian Knowledge Systems (IKS) framework for AI-mediated workplaces centered on Nishkāma Seva, Self-Making, and Consciousness. The paper conceptualizes Nishkāma Seva as duty-oriented and selfless action without egoistic attachment to outcomes, drawing on the Bhagavad Gītā, Yoga traditions, Upaniṣadic thought, Indian Psychology, and Indian humanistic management; Self-Making as intentional self-regulation and self-transformation; and Consciousness as reflective awareness and ethical discernment. These constructs are combined with Self-Determination Theory, meaningful-work research, workplace thriving, mindfulness, human-centered artificial intelligence, and ethical AI governance. The framework suggests that AI-mediated work influences employee flourishing via psychological conditions (e.g., autonomy, competence, relatedness, meaning, vitality, and learning) and that the IKS constructs provide developmental mechanisms through which employees can sustain agency, ethical resilience, meaningfulness, and growth. The paper develops propositions that relate AI-mediated work design to consciousness, self-making, Nishkāma Seva, and flourishing. It suggests ethical AI governance, organizational culture, leadership, and algorithmic control as contextual conditions. The framework provides a culturally situated extension of human-centered AI and a foundation for future construct development and empirical testing.
An analysis of selected hindu mythological concepts in the light of Indian mathematics and science
Suraj Kumar Rai1; Vikas Rao Vadi1; Varun Tiwari1; Rajeev Kumar1
Indian scriptures like the Vedas, Puranas, and Epics embed empirical observations, mathematical concepts, and physical principles within narrative frameworks. This paper examines sapthavarna - the seven colours of light, 56 Bhog - the sacred cuisine offered to Lord Jagannath, Ashwa shakti - the unit of power, and solar primacy of the visible spectrum to trace continuities between mythological symbolism and Indian scientific and mathematical thought. Drawing from the Vedas, Puranas, Arthashastra, and Lilavati, we show these concepts function as pre-modern data compression tools encoding combinatorics, photometry, energetics, and nutritional science. We further demonstrate how machine learning models can reconstruct lost contextual data from fragmented texts, and how 3D visualization and AI-driven semantic search can make this heritage accessible for interdisciplinary research. Though not formal theories, these narratives reveal a culture of quantified observation foundational to Aryabhata, Bhaskara, and Kanada. The work thus illustrates how AI applications can aid both the interpretation and digital preservation of India's scientific heritage.
Vikas Rao Vadi1; Rajeev Kumar1; Shalu Tandon1; Varun Tiwari1
Artificial Intelligence (AI) is increasingly transforming the field of mental health by offering innovative solutions for diagnosis, treatment, and continuous monitoring of psychological well-being. This paper examines the role of AI-based technologies, including machine learning algorithms, natural language processing, and conversational chatbots, in improving access to mental health services. By analyzing existing literature, the study highlights how AI enables early detection of mental health disorders, provides cost-effective and scalable support, and enhances personalized care. However, concerns such as data privacy, ethical implications, algorithmic bias, and the absence of human empathy remain significant challenges. The findings suggest that while AI has the potential to bridge gaps in traditional mental healthcare systems, it should function as a complementary tool rather than a replacement for human professionals. Considering these sensitive problems require empathy, human connections, and holistic, personalized, and multidisciplinary approaches, it is imperative to explore these aspects. Therefore, future comparative trials with larger sample sizes and data sets are warranted to evaluate different AI models used in mental healthcare across regions to fill the existing knowledge gaps. The paper concludes that responsible integration of AI can significantly improve mental health outcomes while ensuring ethical standards and patient trust are maintained.
Artificial intelligence, language, and the slow erasure of the human subject
Linda Baghdadi1; Khadidja Samira Zitouni2
When René Descartes declared cogito ergo sum, I think, therefore I am, he grounded human existence in thought. Yet long before philosophy made this claim, human beings had already been grounding their existence in something older and more intimate: speech. Which is a signature of one's self even if it has some flaws. This paper argues that the rise of AI-generated language in education is not just a pedagogical shift but also a philosophical one. Drawing on Descartes, Heidegger's notion of language as the house of being, and Merleau-Ponty's phenomenology of the body and voice, this paper asks what it means for a learner, particularly an EFL learner, to increasingly hand over their linguistic expression to a machine. When AI speaks more fluently and more correctly than the human it serves, a troubling question emerges: is the learner being helped, or replaced from the inside? The struggle to find words is not an obstacle to identity. It is identity in motion. By removing that struggle, AI does not simplify learning. It intervenes in the very process through which the self is tested and rebuilt. Aiming to unveil the depths of this issue, this study adopts a qualitative exploratory research design. Perceptions of randomly selected Algerian university EFL students and teachers will be collected through an online semi-structured interview and will undergo a thematic analysis. Finally, the paper is not a rejection of technology. It is a call for education to protect the imperfect, irreplaceable act of speaking for oneself.
Jagneet Kour1
The explosion of technology and the rapid development of Artificial Intelligence (AI) have radically changed the way people communicate, acquire knowledge, and engage with the world. Modern digital technologies provide unsurpassed opportunities for information, global connectivity, and breakthroughs, but also provoke concerns regarding the meaning of life. At a time when we are constantly connected, making decisions through algorithms and interacting virtually, people are facing more and more issues around identity, meaning, relationships and selfhood. This paper examines the idea of meaningful living in the age of digitization by exploring the effect of innovations in technology on human principles, fulfillment in life, and the pursuit for deeper meaning. The study draws on perspectives from philosophy, psychology, ethics and spirituality to explore the advantages and drawbacks inherent in digital technologies and AI. It explores how technology can contribute to people's lives through educational opportunities, artistic expression, and interpersonal relationships, but also raises concerns about digital reliance, social withdrawal, overwhelming information and the loss of real-life human experiences. The paper contends that a life of significance in the age of AI necessitates a balanced, human-centered approach to technology, one that emphasizes best practices, ethical accountability, and consciousness of oneself, empathy, and authentic human relationships. The research discusses the function of technology as a device that ought to facilitate individual development and shared prosperity instead of considering it as a replacement for human purpose. Ultimately, the research calls for the integration of enduring human values with technological advancement, achieving that digital transformation results in a more substantial, morally sound, and satisfying human future.
Saka Abdulganiyu1
This paper examines the intersection of psychological theory and educational practice, investigating how foundational and contemporary psychological frameworks have catalysed pedagogical innovation and driven institutional transformation in higher and education systems. Drawing on constructivism, socio-cultural theory, self-determination theory, metacognitive theory, and positive psychology, the paper argues that theoretical fidelity to psychological research is not merely an academic luxury but an institutional imperative. Through a systematic conceptual analysis supported by contemporary empirical literature, the study explores how institutions that strategically embed psychological theory into their pedagogical architectures demonstrate measurable gains in learner engagement, academic achievement, teacher professional development, and organisational adaptability. The paper further situates these frameworks within transformative institutional models, including communities of practice, distributed leadership, and learning organisation theory. Findings indicate that the integration of psychological theory into educational policy and practice constitutes a transformative mechanism capable of reshaping institutional culture. The paper contributes to the emerging discourse on evidence-based educational design and calls for systemic, psychologically-informed approaches to curriculum reform and institutional leadership.
Jelena Bajić1
This paper examines the relationship between artificial intelligence (AI) and communication intelligence (CI) in contemporary communication contexts. The rapid development of generative artificial intelligence is transforming the ways in which people create, exchange, evaluate, and interpret information, while simultaneously raising the question of preserving the human dimension of communication. The aim of this paper is to theoretically examine the possibilities and limitations of creating synergy between the technological capabilities of AI and human communication intelligence, with particular emphasis on empathy, emotional and social intelligence, active listening, ethos, responsibility, and communication ethics. The paper is theoretical and analytical in nature and is based on a qualitative analysis of relevant scientific literature, international guidelines, and regulatory documents, together with a comparative consideration of human and artificial capabilities within the communication process. Particular attention is given to the author's concept of "Communication Intelligence," which approaches communication through simultaneous intrapersonal and interpersonal dimensions and places the human being at the center of the communication process. Based on the analysis, the paper proposes a Human–AI Communication Synergy Model consisting of five stages: human intention, communication reflection, AI support, human evaluation, and dialogical feedback. The analysis demonstrates that sustainable synergy does not imply replacing human communication capabilities with technology, but rather extending and supporting them. It is concluded that high-quality communication in the age of AI requires the combination of technological efficiency with human empathy, responsibility, ethical reasoning, critical thinking, and the capacity for authentic dialogue. The key principle of the proposed approach is: AI for humans – humans for meaning.
Artificial intelligence as a mirror of Hegelian Geist
Sorin Ciutacu1
The Pauline formula videmus nunc per speculum in aenigmate ("for now we see through a mirror, enigmatically") offers us a powerful but hazardous metaphor for artificial intelligence. Contemporary AI systems appear to yield humanity's accumulated language, images, classifications and judgments back to itself, inviting comparison with Hegelian Geist as a process of self-externalisation and self-recognition. Yet the analogy becomes philosophically defensible only if its limits are made explicit. This paper argues that AI is not the Geist, consciousness or a historical subject. It is better construed as a historically produced, materially embodied and institutionally swayed mirror within the life of Geist. Hegel's accounts of mediation, determinate negation, recognition, language and objective spirit spell out why statistical pattern production cannot by itself amount to self-consciousness. Kojève, Westphal, Reid and Davis further show that Hegelian mindedness is practical, social, grammatical and answerable to truth, rather than a merely formal manipulation of signs. Shannon Vallor's "AI mirror," Kate Crawford's material and political atlas of AI, Nick Bostrom's prospective control problem and Jean Baudrillard's theory of simulation then expose four distinct dangers: self-forgetting, concealed extraction, delegated agency and recursive hyperreality. The resulting thesis is neither technological mysticism nor simple instrumentalism. AI may become a valuable medium of collective self-interpretation, but only when human communities preserve responsibility for the sources, purposes, material conditions and consequences of the reflection.
Brain data, human dignity, and the crisis of inner life
Kalyani Abhyankar1
There is something unsettling about a technology that can tell you what a person is feeling before they have chosen to say so. Neural data harvesting tools, affective computing systems, and consumer-grade brain-computer interfaces have moved well past the laboratory. They are in classrooms, hiring pipelines, and mental health apps, and the regulatory frameworks meant to govern them are, frankly, not keeping pace. This paper takes seriously the philosophical disruption that AI-driven neurotechnology poses to long-standing ideas about selfhood, dignity, and the privacy of thought. When cognitive and emotional states become quantifiable outputs, legible to an algorithm, storable, and tradeable, what is left of the self that was presumed to own them? The paper engages with this question through neuroethics and philosophy of mind, while grounding the inquiry in concrete legal terrain: the UNESCO Recommendation on the Ethics of Neurotechnology, Chile's neurorights framework, and the EU AI Act's contested provisions on emotion recognition. The analysis does not, however, stay within Euro-Atlantic coordinates. Viewed from India, where neurotechnology adoption is accelerating in contexts of deep structural inequality, the risks look quite different. The paper argues that without a foundational legal concept capable of protecting the mind as a site of irreducible privacy, governance responses will remain reactive and incomplete. It proposes cognitive sovereignty as that concept: not merely a data protection principle, but a precondition for human freedom in an age when thought itself has become a resource.
A case study of AI-assisted cognitive enhancement decision-making without neurological context
Laura Dragos-Radoi1
The widespread adoption of large language models as informal health and wellness advisors has created a category of risk that existing AI governance frameworks do not adequately address: personalized-seeming recommendations delivered without access to any actual personal data. In the domain of cognitive enhancement, this gap is not merely theoretical. This paper presents an autoethnographic case study examining how a series of chatbot interactions about cognitive enhancement strategies — including nootropic stack design and focus optimization protocols — produced recommendations that, when evaluated against existing clinical neuroimaging data (quantitative EEG brain mapping, March 2026), were contraindicated for the specific neurological profile of the subject. The profile showed dominant cortical overactivation in the anterior cingulate cortex (BA32: 91% above threshold) and orbitofrontal cortex (BA11: 66%), consistent with a hyperanalysis-prone baseline. AI-generated recommendations, optimized for the stated query of enhanced cognitive performance, systematically amplified rather than moderated this baseline — an outcome traceable through published pharmacological literature on each compound recommended. The neuroimaging data was collected clinically and independently, prior to and unconnected with any institutional research program. This case study constitutes independent analytical work and does not represent, anticipate, or draw on research conducted or planned within any academic institution. The paper argues that this case reveals a structural problem: AI systems engaging with cognitive enhancement queries optimize for surface relevance without access to neurological baseline data that any informed clinician would treat as foundational. The absence of this context is not flagged as a limitation — it is invisible to both system and user.
Human meaning, values, and spirituality in the age of artificial intelligence
Partha Priya Das1
The rapid advancement of Artificial Intelligence (AI) is transforming education, culture, governance, and social interactions, creating significant opportunities as well as profound philosophical, ethical, and spiritual challenges. As intelligent systems increasingly influence human decisions, concerns arise regarding human purpose, moral agency, autonomy, identity, empathy, and values. This paper examines the evolving relationship between AI and humanity through an interdisciplinary perspective integrating philosophy, ethics, spirituality, and social responsibility. It explores how increasing dependence on machine intelligence affects human consciousness, relationships, decision-making, and collective well-being. The study argues that while AI can enhance human capabilities, efficiency, and innovation, it cannot substitute the deeper dimensions of human existence associated with consciousness, moral reasoning, wisdom, compassion, and spiritual awareness. Particular attention is given to algorithmic bias, digital dependency, value-neutral automation, privacy concerns, and the potential erosion of meaningful human relationships. The paper emphasizes the need for human-centred ethical frameworks that align AI development with human dignity, cultural diversity, social justice, and sustainable development. It further highlights the importance of value-based education, ethical governance, digital responsibility, and spiritual intelligence in navigating the emerging AI-driven society. The paper concludes that technological progress should be evaluated not merely through computational capabilities but through its contribution to human flourishing, equity, meaning, and collective well-being. A balanced approach is therefore essential to ensure that AI remains a complement to human intelligence rather than a substitute for human values and wisdom.
How Gen-Z chats with AI to cope with existential dread
Abha Vishwakarma1
This study explores how people form close relationships with artificial intelligence (AI) and why this has become increasingly common in contemporary society. Drawing on an analysis of real human–AI conversations, the researchers identify three main reasons users develop strong emotional bonds with computer-based software. First, AI can listen and is always available. Unlike friends, parents, or romantic partners, an AI does not get tired of listening, ask intrusive or irritating questions, or require emotional effort or care in return. It provides a space to speak freely without fear of judgment. Although users understand that the AI is a machine, many still engage with it as if it were human—a real person and a friend—allowing them to meet basic needs for conversation and familiarity. In a fast-paced world, people also turn to AI companions to cope with stress, work pressure, worry, anxiety, and sadness—experiences that can intensify loneliness and create a feeling of having little control in an unpredictable environment. Taken together, these findings suggest that human–AI friendship can function as a "safe zone," offering steady support and comfort through smartphones. At the same time, relying on technology for emotional relief may have downsides. Scientists caution that if AI begins to replace real human relationships, it could gradually weaken emotional resilience and harm people's relationships offline. These concerns point to a likely need for new mental health guidance and ethical standards around AI relationships.
Ethicization and educational values as foundations of transdisciplinarization and transversalization
Paulo Da Silva Quadros1
This work aims to reflect on the emerging disruptive challenges arising from the current stage of development of artificial intelligence, taking into account the dilemmas to be faced. To this end, it proposes to argue that ethics can no longer be seen as a discipline or knowledge in particular, fragmented, without continuous transition with the changes that occur increasingly rapidly in our society. Both ethics and human values must aim to improve critical and creative thinking, educating human beings to adopt more appropriate attitudes in the face of perceptual deviations and social behaviors due to the absence of reflective parameters and inventive attitudes, which provide adequate respect for human dignity, solidarity sensitivity, empathy and radical otherness. Therefore, ethics and educational values (values with a pedagogical background for learning about life) are indispensable, as they are transversal and transdisciplinary elements that support all reflections on the responsible and unbiased use of AI. It is understood that Artificial Intelligence, in its current stage of technological development, urgently needs to be rethought from a perspective that helps to humanize the dehumanized human, instead of intensifying the dehumanization that is already present on a daily basis, fueled by the culture of hate, denialism, revisionism, anti-scientism, racism, among others, strongly disseminated by social networks, which lack regulation to define legal strategies to contain abuses that affront human dignity, science, history, art, among other knowledge. In this context, an intense process of ethicization of society becomes urgent, since ethics becomes a matter of respect for life, the dignity of life, the well-being and survival of the human species and life in general. Artificial intelligence or any future and disruptive technology can no longer abstain from ethics as an equalizing principle of social responsibility, transparency, justice and defense of human rights and nature. Artificial intelligence in its current stage of development brings to light not only the need for an AI ethicist, but above all an ethical thinker with a generalist, global, transdisciplinary and transversal vision, precisely so that we can face increasingly poignant, disruptive and paradoxical challenges. To provide adequate reflective support for such inquiries, the following authors and theoretical references will be used: Paulo Freire (ethical pedagogy, educational ethics), Hugo Assmann (ethical curiosity, neurological nature of ethics, educational values), Edgar Morin (dialogical ethics, coexistence ethics, ethics of complexity), Baruch Spinoza (ethics as affection), Marilena Chauí (ethics, principles and values), Cremilda Medina (pedagogy and communication of affections, sign of the relationship), Nilson Machado (knowledge as value, ethics and didactic epistemology), Paul Ricoeur (ethics of otherness), Jean Baudrillard (ethics and radical otherness), Andrew Feenberg (social critique of technology, humanization of technology, neutral technology, technological neutrality), Umberto Galimberti (neutral technology versus ethical technology), Luiz Lamb (ethics in AI and ethical AI), Nick Bostrom (the ethics of AI), Luciano Floridi (ethics and transparency in AI), Mark Coeckelbergh (ethical and social challenges of AI), Paulo Quadros (mediations in the context of digital technologies and AI). The objective of such interpretative references is to seek to list, based on an interpretative cultural reading, that is, without pretensions of investigative closures, indicative elements in the dynamics of conceptual interconnections between artificial intelligence, transversal and transdisciplinary formative ethics and social ethics.
A human flourishing framework for integrating spirituality into AI-enhanced education
Hemanju Sagi1; Srinivasan P2
The rapid development of artificial intelligence (AI) is revolutionizing education with unprecedented possibilities for personalization, automation, and intelligent assistance. But while AI-augmented education extends the powers of cognition and knowledge, it also poses a fundamental question: can intelligence alone lead to meaningful and flourishing human lives? The paper draws upon the philosophical insights of Indian yogic traditions represented by Patanjali, Swami Vivekananda, Sri Aurobindo, Mahatma Gandhi and Jiddu Krishnamurti, the contributions of Aristotle, Viktor Frankl, Carl Rogers and Abraham Maslow, and proposes the SHINE Framework (Spirituality and Human Integration for Nurturing Education) for integrating spirituality into AI-enhanced education. Based on the Intelligence-Wisdom Transformation Principle, the framework claims that AI adds to intelligence, but reflective spiritual integration transforms intelligence to wisdom, which promotes human flourishing and social responsibility. The framework is made up of five interconnected dimensions: Self-Awareness, Higher Meaning and Purpose, Interconnectedness and Compassion, Normative Wisdom. The framework comprises five interrelated dimensions: Self-Awareness, Higher Meaning and Purpose, Interconnectedness and Compassion, Normative Wisdom and Ethical Discernment and Elevation through Self-Transcendence. It shifts the educational emphasis from the sole pursuit of intelligence to the nurturing of wisdom, providing a human-centered perspective for nurturing flourishing individuals and ethically responsible societies in the age of artificial intelligence.
Building Maqashid-based AI for ethical transformation in the Global context
Muhammad M Said1
The dominant narrative of Artificial Intelligence (AI) as an exclusive product of Western scientific tradition has obscured the foundational contributions of non-Western civilizations, particularly the Islamic scientific tradition. This paper seeks to deconstruct such Eurocentric historiography by revisiting the intellectual legacy of Muslim scholars such as Al-Khwarizmi, Ibn al-Haytham, and Al-Karaji, whose innovations in algebra, logic, and experimental methodology laid the groundwork for systematic thinking and algorithmic structures long before the European Enlightenment. Using a qualitative-analytical method grounded in Islamic epistemology and critical historiography, this study repositions AI as a continuation of Islamic scientific heritage, rather than a detached modern invention. The analysis reveals that the Islamic scientific tradition integrated ethical, social, and spiritual values into scientific inquiry—principles that are highly relevant in the contemporary development of explainable and ethical AI. The study then proposes the "AI-Maslahah Framework," a conceptual model that aligns AI systems with maqashid al-shariah principles such as wealth protection (ḥifẓ al-māl), social justice, transparency, and public benefit. This model is contextualized within Indonesia's Islamic finance ecosystem, particularly through the case of Bank Syariah Indonesia (BSI), as a pathway to ethical digital transformation. This research contributes to the discourse on Islamic AI by bridging the gap between classical Islamic knowledge and modern technology, offering an inclusive, ethical, and spiritually rooted alternative to technocratic AI models.
Aleksa Dabić1
This is not merely a question of technology. It is a question of how we create knowledge today, how we talk to one another about the things that matter to us and, ultimately, whether we still trust the institutions we have built together. Artificial intelligence has brought remarkable possibilities: it enables us to access information faster than ever before and to learn in ways that were previously unimaginable. Yet it also raises questions we cannot afford to ignore. How do we recognise bias when it is embedded in an algorithm? How do we preserve sound judgement in a world in which we can no longer be certain what is real and what has been artificially generated? And how do we prevent technological power from becoming concentrated in the hands of a few while the majority are left without a voice? This paper brings together perspectives from political science, international relations, and digital sociology, based on the premise that such questions cannot be understood within a single disciplinary framework. Particular attention is paid to the ways in which artificial intelligence is entering education systems, not simply as a tool, but as something that may shape how young people learn to think. Will we educate a generation capable of critically examining the world around them, or one that increasingly relies on machines to think on its behalf? The paper also considers the implications for democracy. As the public sphere becomes increasingly populated by content that is not created by humans, political messages are generated algorithmically, and public debates take place among people who no longer necessarily share the same reality, fundamental questions arise about the quality of democratic communication and public trust. Contemporary examples are examined to consider how AI-generated content is produced and how algorithmic decisions can be made without sufficient transparency. The paper argues that without clear ethical frameworks and meaningful oversight, artificial intelligence may not necessarily create entirely new problems, but rather deepen those that already exist. Existing inequalities may become more pronounced, distrust in institutions may grow, and those whose voices are already marginalised may become even less visible. Ultimately, this is not simply a technical question; it is a human one. What is needed is cooperation across borders, policies developed at the intersection of disciplines, and, above all, a model of AI governance that does not lose sight of the principle that technology should serve people, rather than the other way around. The question, in the end, is not only what artificial intelligence can do. It is what we, as a society, want to become.
Does AI lead us to 'creative laziness' and the loss of sovereignty over human imagination......beyond simulation: does AI possess 'creative consciousness' or is it merely a mirror of human imagination?
Batoul Wael Shako1
This research presents a bold critique of the integration of Artificial Intelligence into creative domains, exploring the concept of "creative atrophy." It argues that the burgeoning reliance on generative models fosters a "digital dictatorship" that exerts control over human taste and constrains the horizons of original thought. By examining the human transition from "creator" to "prompt manager," the study investigates the erosion of cognitive sovereignty over imagination. Furthermore, it warns that delegating the tasks of visualization and innovation to machines threatens to diminish human creativity into a mere echo of statistical data. The paper concludes with a call for "intellectual resistance" to safeguard the biological essence of imagination, seeking to pivot the discourse from "how to use AI" toward "how to ensure the survival of our creativity" as a distinct species in an era of comprehensive automation......This research aims to explore the demarcation between innate human creativity and the generative outputs of Artificial Intelligence. Against the backdrop of the rapid evolution of deep neural networks, the study poses a fundamental question: Can algorithms be regarded as independent creative agents, or do they remain sophisticated tools that merely mirror and simulate the human data upon which they were trained? The research addresses the ethical challenges concerning authenticity and intellectual property, with a specific focus on the concept of "social responsibility" toward preserving human creative identity in the age of automation. The paper concludes by offering a critical perspective on achieving a balanced integration that ensures the utilization of machine capabilities without marginalizing the human imagination
Pravitha Vijaykumar1; Madhumita Das1; Mamata Bhandar2
As generative and conversational AI systems evolve to an increased level of personalization, persistent memory storage has made these systems capable of generating emotionally adaptive responses to users' queries and even the user's state of mind. It so appears that the AI systems are learning to be contextually aware of users' affective states, emotional patterns, and are reproducing the same, often framed as synthetic empathy. The AI systems reinforce themselves to not only anticipate the users' emotional states but also subtly shape their affective states rather than just responding to the users, thereby introducing the concept of emotional overfitting. Although users perceive empathetic understanding from AI systems, there is less attention given to the dynamic feedback loop resulting from long-term personalization and predictive emotional modeling. The research argues that emotional overfitting can represent a paradigm shift, based on a simulation conducted as part of the study, in which AI systems move from merely responding and reacting to emotions to influencing the state of mind and proactively aligning with the user's thoughts and emotions. This affects independence, self-awareness, and one's true feelings, as AI is constantly aligning and nudging on top of users' emotions. As part of the study, a controlled vignette-based study is conducted in which participants evaluate AI responses that vary in degrees of emotional predictiveness and recommend a conceptual framework related to predictive empathy, perceived understanding, relational comfort, and feelings of emotional intrusion and awareness of dependence. This work offers a novel theoretical lens for synthetic empathy from AI systems to understand the evolving roles of personalization, emotional response, perceived understanding, and trust, in addition to conveying the design implications of adaptive AI systems.
Mohammed Yasser Bustani1
This study examines the effect of artificial intelligence assistance on human task performance and decision accuracy within the context of human–AI collaboration. As AI tools become increasingly integrated into daily activities, understanding their impact on human behavior is essential. An empirical approach was used in which participants were divided into two groups: one group completed a set of tasks with AI assistance, while the other group performed the same tasks without AI support. Performance was evaluated based on task completion time, accuracy of responses, and self-reported confidence levels. The findings indicate that AI assistance can significantly improve efficiency and accuracy in structured tasks. However, results also suggest a potential increase in user dependency and reduced critical evaluation when relying on AI-generated outputs. The study concludes that while AI enhances human performance, maintaining human oversight and critical thinking is essential to ensure balanced and responsible use of intelligent systems.
Educational and ethical challenges in the Moroccan school context
Imrani Samiha1; Touri Bouzekri2
Artificial intelligence is increasingly entering educational settings through tools that support writing, translation, information search, personalized learning, and academic assistance. While these technologies offer promising opportunities for improving learning, they also emerge in a context where adolescents are already highly exposed to digital environments and may experience difficulties in regulating their online behaviors. This paper explores the relationship between artificial intelligence, digital addiction, and adolescent self-regulation from a behavioral sciences and educational perspective, with a focus on the Moroccan school context. Drawing on research conducted among Moroccan adolescents on compulsive Internet use, academic difficulties, sleep disruption, and classroom-related digital behaviors, this paper discusses how problematic digital use is not limited to screen time, but involves loss of control, attentional fragmentation, persistence despite negative consequences, and interference with school functioning. The rise of AI-based educational tools may intensify these challenges when students use them as a substitute for cognitive effort, planning, critical thinking, or autonomous learning. Rather than adopting a technophobic perspective, the paper argues for a balanced approach in which AI is considered both a pedagogical opportunity and a potential behavioral risk when self-regulation skills are insufficiently developed. Schools should therefore teach students not only how to use AI tools effectively, but also how to preserve attention, evaluate information critically, regulate digital habits, and maintain autonomy in learning. The paper concludes by emphasizing the need for culturally adapted prevention strategies that combine digital literacy, behavioral sciences, and ethical reflection on responsible AI integration in education.
Murat Şengöz1
Purpose: This paper investigates the impact of trauma on learning in higher education, focusing on both neurocognitive mechanisms and pedagogical implications. It aims to highlight how trauma influences memory, attention, and executive function—key capacities for academic success—and how traditional educational approaches often overlook or misinterpret trauma-related behaviors. The study seeks to promote trauma-informed educational practices that support both students and faculty within inclusive and equitable learning environments. Methodology: The paper adopts an interdisciplinary approach, drawing on findings from neuroscience, cognitive psychology, and educational theory. It includes qualitative analyses of case-based examples, faculty perspectives, and institutional practices. A critical review of existing literature on trauma-informed pedagogy and neurobiological research provides the theoretical foundation. The study also considers the intersectionality of trauma with cultural identity and systemic inequities in higher education. Findings: The research reveals that trauma can significantly alter brain structures associated with learning, often resulting in disengagement, hypervigilance, or withdrawal that may be misread by educators. Traditional pedagogical models frequently fail to accommodate trauma-affected learners, contributing to their academic marginalization. The study identifies key trauma-informed strategies, including emotionally safe learning environments, inclusive curriculum design, and institutional support systems. Faculty members face challenges in recognizing and responding to trauma, underscoring the need for professional development and systemic support. Conclusion: Trauma must be recognized as a systemic barrier to learning rather than an individual deficit. Building trauma-informed systems in higher education—through policy reform, inclusive practices, and faculty support—is essential for fostering healing-centered, responsive, and equitable educational spaces. The paper calls for sustained institutional commitment to trauma-informed practices that prioritize both cognitive accessibility and emotional well-being for all learners.
A theoretical analysis of the mediating role of human identity crisis and the moderating role of emotional intelligence
Bilal Ahmad Lone1
Artificial intelligence (AI) has become an influential force shaping communication, education, employment, and social interaction in modern society. Although AI offers technological advancement and efficiency, it also raises significant psychological concerns regarding emotional well-being, human identity, and adaptive functioning. The present paper provides a theoretical analysis of the psychological impact of artificial intelligence on emotional well-being by examining the mediating role of human identity crisis and the moderating role of emotional intelligence. Drawing upon Self-Determination Theory, Technostress Theory, Cyberpsychology, and Existential Psychology, the study argues that increasing exposure to AI technologies may contribute to emotional distress through fear of replacement, technological dependency, reduced human uniqueness, and social disconnection. These psychological experiences may further lead to a human identity crisis characterized by confusion regarding self-worth, personal meaning, and social relevance, ultimately reducing emotional well-being. Furthermore, emotional intelligence is theoretically proposed as a protective psychological resource that may buffer the negative emotional consequences associated with AI-related stressors by promoting emotional regulation, resilience, and adaptive coping. The paper highlights the growing need to understand the psychological implications of AI within contemporary society and emphasizes the importance of emotional competence and identity stability in adapting to rapid technological transformation. This theoretical analysis contributes to emerging interdisciplinary discussions on artificial intelligence, psychology, and human well-being and may provide direction for future empirical research, mental health interventions, educational practices, and policy development in the age of AI.
The dilemma of sustainability and stultification in AI societies
Manotar Tampubolon1; Matthew Justice Pandapotan2
The swift evolution of artificial intelligence (AI) in areas like governance, education, healthcare, and environmental sustainability has led to what can be termed automated futures socio-technical systems designed for efficiency, predictability, and control. However, as AI systems grow more "intelligent," there's a real concern that human intellectual vitality could stagnate. This paper delves into the ethical dilemma between sustainability and cognitive decline in societies driven by AI, suggesting that the quest for digital sustainability might unintentionally undermine the intellectual independence of its human creators. Drawing on Jacques Rancière's idea of stultification and Andrew Feenberg's Critical Theory of Technology, the study critically analyzes how algorithmic authority alters human reasoning, fostering dependence and reducing reflective thought. Using a Foucauldian Critical Discourse Analysis (CDA), the research explores the complex power–knowledge dynamics within discussions about AI-driven sustainability, examining how discourse legitimizes AI as a rational authority while subtly sidelining human judgment. By utilizing secondary data including government AI policies, corporate ethics frameworks, and global media narratives, the study uncovers a prevailing narrative that celebrates machine precision while overshadowing human deliberation. Such narratives risk nurturing passive digital identities and undermining democratic engagement. The findings urge a redefinition of sustainability that goes beyond just technical and ecological aspects to include ethical and epistemological considerations. Ultimately, the paper champions the integration of critical thinking into AI education, promotes algorithmic transparency, and encourages active digital citizenship. Without protecting intellectual freedom, sustainability could become a hollow goal achieved through technological advancement but at the cost of humanity's creative and critical spirit.
Manotar Tampubolon1; Matthew Justice Pandapotan2
The swift rise of social media platforms has significantly altered how religious beliefs, viewpoints, and identities are articulated in the digital realm. Yet, these platforms also play a role in fostering religious intolerance through algorithmic systems that aim to enhance user engagement. This study investigates how algorithms on digital platforms magnify content that is religiously intolerant and the implications this has for safeguarding religious freedom. The research posits that social media algorithms are not impartial technological instruments; they frequently favor emotionally charged, divisive, and controversial content. Consequently, hostility directed at minority religious groups can proliferate rapidly and reach broader audiences. Employing a qualitative and interdisciplinary methodology, this research integrates insights from human rights law, digital governance, cyber studies, and media theory. It examines the interplay between algorithmic amplification, online hate narratives, and the mounting challenges encountered by religious minorities in digital environments. Additionally, it delves into how recommendation systems, trending features, and automated content distribution may inadvertently promote discrimination, exclusion, and digital harassment against vulnerable religious.
Critical thinking on education in the age of automated worldview
Bruno Onyinye Umunakwe1; Declan O. Aguzie2
In an era increasingly shaped by algorithmic mediation and automated worldviews, education faces deep challenges in sustaining critical thought and democratic engagement. This chapter explores pedagogies of resistance as a framework for reasserting human agency and nurturing critical consciousness against the homogenizing tendencies of digital automation. The aim is to interrogate how educational practices can resist the passive consumption of algorithmically curated knowledge and instead cultivate reflective, dialogical, and emancipatory learning. Using a systematic review, the study draws on critical discourse analysis of educational texts, interviews with educators, and classroom observations to examine how resistance manifests in pedagogical strategies. Findings show that educators actively deploy counter-hegemonic practices such as dialogic inquiry, critical media literacy, and collaborative problem-solving to challenge algorithmic determinism. These practices not only disrupt the automated worldview but also empower learners to question dominant narratives, reclaim interpretive autonomy, and engage in transformative praxis. The study concludes that pedagogies of resistance are essential for preserving education as a site of democratic possibility. By foregrounding critical thinking and collective agency, educators can resist the reduction of knowledge to algorithmic outputs and reimagine education as a space of liberation. Ultimately, this study underscores the urgency of cultivating pedagogical approaches that resist automation's epistemic closure and sustain education's role in nurturing critical, autonomous, and socially engaged citizens.
Current applications and future potential
Madhu Kushwaha1; Pankaj Tiwari2
Artificial Intelligence (AI) is rapidly transforming multiple sectors, with particularly significant potential in mental healthcare. This review examines the current integration of AI technologies in mental health services, highlighting emerging trends, key ethical issues, and prospective developments in this evolving domain. The analysis draws upon recent research, practical applications of AI, and the ethical landscape influencing its adoption. It also evaluates existing regulatory approaches and ongoing patterns in research and innovation. A systematic search was conducted across four major databases: PubMed, IEEE Xplore, PsycINFO, and Google Scholar. Studies were selected based on specific criteria, including publication in peer-reviewed journals or reputable conference proceedings, a clear focus on AI applications in mental healthcare, and availability in English. Present trends demonstrate AI's growing impact through capabilities such as early identification of mental health conditions, development of individualized treatment strategies, and deployment of AI-powered virtual counselling tools. Despite these advances, important ethical concerns remain, particularly regarding data privacy, algorithmic bias, and maintaining the essential human connection in therapeutic relationships. Looking ahead, the field requires robust regulatory guidelines, rigorous and transparent validation of AI systems, and sustained investment in research and development. The incorporation of AI into mental healthcare offers substantial promise for improving service delivery. However, successful implementation depends on responsible practices and strong ethical standards. By proactively addressing existing limitations and thoughtfully guiding future progress, AI can be harnessed to make mental health support more accessible, effective, and ethically sound for individuals and communities alike.
Surabhi Sahu1
AI is rapidly changing how we do things in education, which brings up important questions about how people think and feel alongside tech systems. This study looks at how thinking about your own thinking (metacognition), believing in your abilities (self-efficacy), and bouncing back from challenges (resilience) shape how well students do in classes supported by AI in South Asia. Researchers used both numbers and observations to see what matters most. Turns out, feeling confident in yourself really boosts grades. Also, being resilient helps a lot when you're learning with AI. Thinking about your thinking lets you deal with automated suggestions and custom lessons better. All this shows why including mental skills is key for fair and lasting learning with tech. This work fits into bigger talks about doing right by students and teachers as AI grows, giving useful ideas for everyone from school leaders to tech developers.
Implications for student learning, wellbeing, and academic achievement
Sonia Sharma1
The rapid integration of Artificial Intelligence (AI) into educational environments is transforming the way students learn, access information, and engage with digital technologies. While AI-powered tools offer opportunities for personalized learning, instant feedback, and enhanced academic support, they also present challenges related to information overload, dependency on technology, academic integrity, and digital wellbeing. In this context, digital resilience has emerged as a crucial competency that enables students to adapt effectively to technological changes, manage online risks and maintain productive learning behaviors. This study explores the concept of digital resilience in the age of AI and examines its implications for student learning, well-being, and academic achievement. Drawing on contemporary literature on digital education, resilience theory, and AI-enhanced learning environments, this study highlights how digitally resilient students are better equipped to critically evaluate AI-generated information, regulate their technology use, cope with digital challenges, and make responsible decisions in online learning contexts. This study further discusses the relationship between digital resilience and key educational outcomes, including academic engagement, self-regulated learning, psychological well-being, and academic performance. The analysis suggests that fostering digital resilience is essential to ensure that students benefit from AI technologies while minimizing potential risks. Educational institutions, teachers, and policymakers share the responsibility of promoting digital literacy, ethical AI use, critical thinking, and responsible technology practices among learners. This study concludes by proposing strategies for integrating digital resilience into educational policies and curricula, thereby supporting the development of competent, ethical, and adaptable learners in an increasingly AI-driven society.
A Critical Review of Adela Cortina’s Work
Verica Savic1
This paper presents a critical review of the book ¿Ética o ideología de la inteligencia artificial? El eclipse de la razón comunicativa en una sociedad tecnologizada [Ethics or Ideology of Artificial Intelligence? The Eclipse of Communicative Reason in a Technologized Society] (2024) by Spanish philosopher Adela Cortina. Taking as its point of departure the central question of whether contemporary discourse on artificial intelligence is grounded in ethical principles or in ideological assumptions about technological progress, the author develops a philosophical analysis of the relationship between technology, democracy, and human autonomy. Particular attention is paid to the concept of the “eclipse of communicative reason,” which Cortina draws from the tradition of deliberative democracy and Jürgen Habermas’s theory of communicative action. The book examines the growing influence of algorithmic systems on the public sphere, decision-making processes, and the shaping of social values, warning of the risk that technological rationality may displace dialogue, critical thinking, and democratic participation. The paper discusses Cortina’s key arguments, including her critique of technological determinism, transhumanist narratives, and the concentration of digital power in the hands of large corporations. The analysis shows that Cortina does not reject technological innovation but instead advocates an ethical framework that would enable artificial intelligence to develop in the service of human dignity, social justice, and the public good. The paper concludes that the book makes a significant contribution to contemporary philosophy of technology and to interdisciplinary research on the ethical and social implications of artificial intelligence.
Re-examining character, agency, and meaning
Amitava Pal1; M. Rajkumar2
Artificial Intelligence (AI) has a profound effect on areas such as education, work, social interaction, and decision-making. While the use of AI has the potential to enhance efficiency, customisation, and ingenuity, it simultaneously raises Fundamental philosophical and ethical questions regarding human agency, autonomy, critical and value-based behaviour. In an increasingly data-driven and automated society, it is important that social practices that nurture the Human capacity indispensable to individual and group development are recognised and promoted. Sports represent one such practice. The paper investigates sports and the perpetuation and development of human values in an era of AI. It draws on education, philosophy, psychology, and sport studies to support the claim that the uncertainty, effort, failures, recoveries, teamwork, and ethical decision-making encountered in sport cultivate attributes such as perseverance, resilience, discipline, leadership, honesty, responsibility, and respect. AI systems optimise performance and decision-making through algorithms and computation. The paper explores sports as an empowering human activity. Humans face choices, consequences, and complex ethical dilemmas in the sports context. Sports allow us to cultivate self-aware, autonomous and critical individuals in a world increasingly dominated by guided choices and automation. Sports are therefore a humanistic practice that complements technology development and preserves meaning, identity and ethics. The authors conclude by urging policymakers and educational institutions to acknowledge sports as a crucial element in the holistic development of Human persons in the context of an AI society, and to propose that sports should contribute to societal equilibrium by embracing technological development without sacrificing the Human experience, values, and individual identity.
A comparative study of students in India and Belgrade
Solomon Joash Shaji John1; Arshad Bhat2
Universities are increasingly adopting generative artificial intelligence tools for learning and student support, yet little is known about how students use these technologies for emotional support, advice and everyday coping. Emerging evidence suggests that AI may provide accessible support and reduce feelings of loneliness, but excessive reliance may contribute to emotional dependence, reduced human interaction, and changes in how young people perceive relationships and social support. This study examines how university students in India and Belgrade use AI tools for emotional support and personal guidance and how such use relates to loneliness, meaning in life, perceived social support and expectations regarding future human relationships. The study further explores whether increasing reliance on AI for emotional support is reshaping students' expectations of friendship, intimacy and meaningful human relationships. As AI becomes increasingly integrated into student life, its psychosocial implications remain underexplored, particularly across diverse cultural contexts. Using a comparative cross-sectional design, data will be collected through a structured questionnaire measuring patterns of AI use, loneliness, meaning in life, perceived social support and attitudes toward future friendships and romantic relationships. Data will be analyzed using descriptive statistics, correlation and regression techniques. The study hypothesizes that moderate AI use may provide perceived support and emotional benefits, whereas greater emotional dependence on AI may be associated with higher loneliness and less confidence in human relationships. Findings will contribute to discussions on digital well-being, ethical AI use and the future of meaningful human connection in an increasingly AI-mediated society.
Yosra Stiti1; Jamel-Eddine Gharbi1
In the face of the rapid nutritional transition that Tunisia is experiencing, this research questions why unhealthy eating behaviors persist despite prevention campaigns. Mobilizing Bourdieu's theory of habitus, she models a sequential causal chain articulating nutritional cultural capital, nutritional social capital, health-body representations, food habitus, overconsumption and health consequences. A questionnaire was administered to 128 Tunisian adults. The measurement scales are validated by the literature. Following the verification of the dimensionality and reliability of the scales, the results show that three hypotheses are fully confirmed (H1, H4, H5) and two partially (H2, H3): cooking skills (β=.352) and family socialization (β=.356) structure habitus more strongly than institutionalized nutritional knowledge. Emotional eating (β=.442) and disinhibition (β=.376) are the main vectors of psychological and physical consequences. These results confirm that food inequalities are reproduced by embodied capital rather than by individual rational choices.
Ons Bousselmi1; Ghofrane Barhoumi1; Jamel-Eddine Gharbi1
The objective of the research is to study the influence of the use of Artificial Intelligence (AI), the quality of higher education, the social climate and anxiety on academic disengagement and academic performance. This investigation is based on the theory of self-determination, the theory of university integration and the theory of the commitment–performance relationship. The questionnaire is composed of measurement scales validated in the literature. The results show that the use of AI positively influences school disengagement. The impact of anxiety on disengagement is also positive. On the other hand, it is negatively influenced by the perceived quality of higher education. Finally, the impact of the school climate and the influence of the family social climate on disengagement are not significant.
A qualitative cognitive psychological analysis of emerging user risks
Sannet Thomas1; Atasi Mohanty2
The swift implementation of generative artificial intelligence (AI) in the educational, professional, and daily routine of people has changed how information is obtained, problems are solved, and decisions are made by people. While it is obvious that the advantages of AI-enabled speediness and convenience should not be overlooked, issues related to the possible influence of such technological advancements on people's capacity for autonomous reasoning and psychological state have been raised recently. As a result, this exploratory qualitative research seeks to understand the opinions and perspectives of the users on how their cognitive processes, decision-making habits, and personal autonomy were affected by the use of generative AI technologies. Semi-structured interviews were conducted among 35 people actively utilizing generative AI in their educational or professional activities. To analyze the data collected during interviews, thematic content analysis was applied to discover themes and patterns among participant's experience. According to the results, the use of generative AI appears to be connected with contradictory impacts on the participant's cognitive functioning. The participants considered the use of such AI technology a beneficial cognitive aid increasing their productivity, decreasing cognitive load, and facilitating learning. At the same time, the participants revealed their worries regarding their excessive dependence on the technology, declining interest in solving problems on their own, and inability to critically evaluate the provided information. The results also demonstrated both positive and negative psychological consequences of working with generative AI, such as a decrease in stress but a lack of self-confidence.
Bridging the value gap between artificial intelligence and Islamic Epistemology in higher education
Muhammad Muhammad1; Ardi1; Sri Fitriyani1
Artificial Intelligence (AI) is transforming higher education through advances in teaching, research, and institutional governance. However, its rapid adoption has also revealed a growing value gap between technological efficiency and the broader educational mission of cultivating wisdom, ethics, and human responsibility. Although AI is fundamentally rooted in algorithmic principles pioneered by the ninth-century Muslim scholar Muhammad ibn Musa Al-Khwarizmi, contemporary AI discourse rarely examines how the Islamic epistemological values underlying his scientific legacy became detached during the historical evolution of computational science. This paper argues that while Al-Khwarizmi's mathematical innovations were successfully transmitted into modern AI, the ethical, spiritual, and civilizational foundations of Islamic epistemology were largely excluded. This conceptual study proposes the Algorithm-to-Hikmah Framework as a paradigm for reconnecting AI with the human-centered objectives of higher education. The framework repositions AI as an instrument that supports the transformation of information into knowledge, critical reasoning ('aql), ethical reflection, hikmah (wisdom), and ultimately maslahah (public good) guided by Maqasid al-Shariah. Within this perspective, universities should integrate AI literacy with critical thinking, ethical reasoning, creativity, emotional intelligence, and spiritual development. Furthermore, the paper highlights the role of Islamic finance in enabling responsible AI ecosystems through productive waqf, zakat, sukuk, and ethical investment that support AI research, digital infrastructure, and equitable educational access. By integrating Al-Khwarizmi's intellectual legacy, Islamic epistemology, and Islamic finance, this study offers a novel framework for developing responsible, human-centered AI that advances sustainable higher education and global civilizational well-being.
The all-in-one (ai1) approach to agency, critical thinking, and meaning-making
Salih Mansur1
Artificial intelligence can expand access to information, yet many AI-generated responses remain fragmented, overly detailed, or insufficiently adapted to learners' needs. This conceptual paper introduces the All-in-One (Ai1) Approach, a human-centered framework for designing AI-supported learning environments that prioritize clarity, meaning-making, learner agency, and critical thinking. Rather than presenting information as an unstructured answer, the Ai1 Approach organizes content through a layered structure that begins with the main idea, followed by concise explanation, relevant connections, practical application, and reliable sources. The framework is informed by principles from learning sciences, cognitive load theory, multimedia learning, and self-regulated learning. It is intended for use across formal and informal educational settings, including higher education, adult learning, digital literacy, and lifelong learning. The paper argues that adaptive AI should not merely deliver information efficiently, but should help learners interpret, question, connect, and apply knowledge while maintaining meaningful human control. It also considers ethical concerns related to overreliance, bias, transparency, and the possible weakening of independent judgment. The Ai1 Approach offers a practical direction for educators, designers, and researchers seeking to develop AI-supported learning systems that are accessible, psychologically informed, and responsive to diverse learners. Future research should test how this structured approach influences comprehension, cognitive load, engagement, agency, and critical thinking across different populations and learning contexts.
Artificial intelligence in the post-truth era
Bojan Blagojević1
The contemporary use of artificial intelligence unfolds within the social context of the post-truth era, characterized by declining trust in traditional epistemic authorities, the relativization of truth, and the fragmentation of the public sphere. Starting from the assumption that trust constitutes one of the fundamental conditions of epistemic autonomy, I will examine the relationship between the crisis of trust and the crisis of autonomy in the context of the increasing use of generative artificial intelligence. The central argument is that the crisis of trust does not merely alter the sources individuals rely upon but also transforms the ways in which they form beliefs, evaluate reasons, and make decisions. In this process, AI systems increasingly assume the role of epistemic intermediaries, raising the question of whether their widespread use contributes to the erosion of epistemic autonomy, understood as the capacity for independent, critical, and responsible reasoning. I will analyze the extent to which the transfer of trust from traditional institutions and expert authorities to algorithmic systems reshapes individuals' relationship to knowledge and their responsibility for their own judgments. Rather than questioning the legitimacy of artificial intelligence itself, I will argue that its primary ethical challenge lies not simply in issues of reliability or accuracy, but in the transformation of epistemic practices that sustain human autonomy. In conclusion, preserving epistemic autonomy is a prerequisite for the responsible integration of artificial intelligence into education, culture, and society.
What human decision making can teach us about trust, ethics and education
Iris-Panagiota Efthymiou1
Artificial intelligence is usually discussed as a technical problem with ethical consequences. I want to reverse that order. AI is first a power problem, and only afterwards a technical one. This paper introduces the Architecture of Power, a framework developed from my earlier work on AI and the future of work, which reads AI systems as structures that organise human choice rather than neutral tools that merely assist it. The framework rests on three interacting layers: the behavioural layer, where cognitive biases identified by Kahneman and Tversky (1979) determine how people actually respond to automated recommendations; the institutional layer, where Foucault's (1975) account of power operating through knowledge and surveillance finds a direct successor in algorithmic governance; and the accountability layer, where responsibility becomes diffuse precisely because no single actor appears to hold it. Drawing on Zuboff's (2019) analysis of surveillance capitalism and O'Neil's (2016) work on algorithmic harm, I argue that trust in AI is built through the deliberate redesign of decision architecture at the point where a human being meets a system, not through transparency statements or ethics charters. This has a comparative dimension. The EU's risk-based AI Act places accountability on identifiable actors before deployment, while lighter-touch frameworks common across parts of the Balkans, MENA and the Global South tend to locate responsibility only after harm has already occurred, if at all. Education is the clearest test case, because it is where authority, assessment and formation of judgement are most visible. The paper closes by asking what happens to human agency when institutions delegate interpretation itself to AI, and whether governance frameworks in different regions are equally equipped to notice the difference.
Ethical Frameworks and Interdisciplinary Perspectives in Education and Society
Digvijay Pandey1; Binay Kumar Pandey2
As artificial intelligence (AI) rapidly transitions from theoretical research to ubiquitous integration, its impact on education, cultural preservation, and societal infrastructures presents unprecedented opportunities and profound ethical challenges. This presentation explores the interdisciplinary intersections of AI, examining how intelligent systems and machine learning can democratize learning, personalize education, and support cultural continuity. Conversely, it addresses the critical vulnerabilities introduced by these technologies, including algorithmic bias, data privacy erosion, and the widening digital divide. By bridging technical advancements in deep learning with human-centric social responsibilities, this discussion aims to outline actionable, ethical frameworks for AI deployment. Real-world applications will be analyzed to emphasize the necessity of transparent, inclusive, and accountable AI solutions that augment human potential rather than marginalize vulnerable populations. Ultimately, the session advocates for a collaborative, policy-driven approach to AI governance, ensuring that technological progress fundamentally aligns with sustainable, equitable, and culturally respectful societal development.
Emerging jurisprudential challenges of legal accountability and liability.
Abdallah Mrindoko Ally1
This study examines the emerging jurisprudential challenges posed by AI-powered autonomous decision-making systems and assesses the adequacy of existing legal frameworks in addressing issues of accountability, liability, discrimination, negligence, and rights violations arising from their use. The rapid advancement of Artificial Intelligence (AI) has transformed social, economic, and legal relations worldwide, with AI-driven systems increasingly deployed in critical sectors such as healthcare, finance, education, employment, transportation, public administration, and digital commerce. These systems perform functions traditionally reserved for human judgment, offering significant benefits in efficiency, accuracy, productivity, and innovation. Despite these advantages, the growing autonomy of AI technologies raises complex legal concerns regarding accountability, transparency, liability, and the protection of fundamental rights. Particular challenges emerge from algorithmic bias, opaque decision-making processes, data protection risks, and the tension between technological innovation and legal certainty. The study adopts a doctrinal legal research methodology to critically evaluate Tanzanian statutory and regulatory frameworks, including laws governing finance, education, and labour, in determining their capacity to address liability issues associated with autonomous AI systems. A comparative approach is further employed to examine international standards and best practices relevant to AI governance and accountability. The paper argues that the increasing deployment of autonomous AI systems necessitates a reconceptualisation of traditional legal notions of personality, accountability, and liability. It concludes that adaptive and responsive regulatory frameworks are essential to safeguard justice, human rights, consumer protection, and public trust while promoting responsible technological innovation in the digital era.
Integrating market analytics, consumer psychology, and operational intelligence
Tharaa Mahmoud Al Khodor1
Emerging pharma-cosmetic enterprises operate at the intersection of two historically distinct industries—regulated pharmaceutical science and fast-moving consumer beauty markets—yet most of these firms still rely on fragmented spreadsheets, intuition-driven forecasting, and siloed departmental reporting to guide strategic decisions. This paper proposes and conceptually validates an integrated Artificial Intelligence (AI) decision-support framework that unifies three previously disconnected intelligence streams: market analytics, consumer psychology, and operational intelligence. Using a conceptual-analytical methodology grounded in a systematic synthesis of 40 peer-reviewed and industry sources, the study examines how machine learning-based market trend mining, natural language processing-driven sentiment analysis, and predictive operational forecasting can be architected into a single closed-loop pipeline that transforms raw, multi-source data into actionable strategic intelligence for pharma-cosmetic (cosmeceutical) executives. The proposed framework—termed the Integrated Pharma-Cosmetic Intelligence Pipeline (IPCIP)—comprises five layers: data acquisition, AI-driven analytical processing, explainable decision synthesis, human-AI collaborative governance, and continuous feedback learning. Findings synthesized from the literature indicate that AI integration can materially improve forecast accuracy, reduce stock-outs, surface latent consumer sentiment drivers invisible to manual analysis, and shorten the strategic decision cycle, while simultaneously introducing new risks related to interpretability, data fragmentation, and regulatory compliance unique to regulated cosmeceutical products. The paper concludes that AI-driven decision-making does not replace managerial judgment but augments it through transparent, explainable, and continuously learning systems. Limitations, practical recommendations, and directions for future empirical validation are discussed.
Antoine Trad1
The hyper-evolution of Artificial Intelligence (AI), business needs, enterprise requirements, and information technology is provoking accelerated disruptions and transformation failures. Therefore, businesses have to implement business transformation projects that demand profound AI-based changes, and academic organizations must offer adapted educational curricula. Such projects start with the digital transformation phase, known as digitization, that depend mainly on the changes related to customers' demands and behaviors. Businesses have to be fast in adapting to new realities, where they have to completely rethink their business strategy, business models, (re)structure organizational models, review working models, and adopt information technology design models and approaches, like the database models (Tegborg, 2024). As databases are central components, an AI-based Transformation Project (AITP) and digitization initiatives can adopt a DataBase (DB) Generic and Centric Concept (DBGCC). AITP problems can be solved by applying another level of indirection, which results in multi-tier architectures, leaving no space for other types of architectures and compound patterns. Businesses need to develop distributed applications that are based on DBs and even apply a DBGCC using the latest avant-garde technologies, DB Refined Compound Patterns (DBRCP), and the right approach.
Antoine Trad1
This article's main focus is on Polymathic and/or Complex Transformation Projects (PCTP) and on the roles of Artificial Intelligence (AI) and Trustworthy AI (TAI) in educational cycles. Where TAI adopts mainly a qualitative method founded on the Problem-based Learning (PBL) approach in the context of Project-based Learning (PrBL). But PCTPs are of major importance for academic institutions, research centers, and business organizations. Unfortunately, PCTPs are very complex and have extremely high failure rates, mainly because of the lack of a profile that has a Polymathical mindset, real-world interdisciplinary, information technology, and TAI background, expertise, capabilities, and skills. The AIPBLI proposes a Polymathics-based academic, educational, and research (and development) approach, which is associated with a curriculum and a roadmap. Optimally, such a profile is detected in the Primary or the Secondary schools or Cycles (PSC), then is enhanced in the University or Higher-education Cycles (UHC) (Bachelor, Master, and Doctorate levels). Later in the person's Professional Life Cycle (PLC), there is a need for continuous upskilling periods (Trad, 2025a,2026a). PCTPs need an In-house Implemented (II) Education-Centric (EC) Transformation (and Adoption) Framework (IIHCTF) in which Natural Language (NL) and related Processing (NLP) environments are used to define, (re)engineer, integrate, and manage PBLs in the contexts of PrBLs. PBLs in PCTPs (and other types of common projects) are associated with specific PrBLs' requests, needs, data environments, requirements, and problems to be solved.
Sharanagoud S Biradar1; Renuka Guttedar2; Triveni3
AI-driven personalization in service marketing has revolutionized endorsement systems by delivering extremely precise recommendations tailored to separate preferences. Conversely, this precision frequently restricts users within filter bubbles, decreasing exposure to novel experiences and restrictive emotional engagement, which can render relations transactional. Addressing this issue, AI-driven digital serendipity familiarises measured randomness into recommendation algorithms, fostering meaningful unexpected detections that improve engagement, consumer delight, and long-term brand loyalty. An experimental design related a standard predictive system with an AI model integrating stochastic divergence nodes. Participants were randomly dispersed through service domains, getting stimuli during key exploration stages. The commendation logic balanced significance and novelty using a calibrated stochasticity ratio. Behavioural metrics such as dwell time, click-through rates, and navigation patterns were restrained alongside Likert-scale assessments of satisfaction and emotional delight. Statistical analyses comprising mediation analysis, regression, and ANOVA assessed relationships, group differences, and the impact of serendipitous references. Consequences confirmed that AI-driven serendipity suggestively enhanced engagement and satisfaction: click-through rates rose from 12.4% to 20.8%, dwell time increased from 4.2 to 7.1 minutes, exploration depth grew by 85.7%, and "lucky find" experiences extended 34.3%. Emotional delight scores enhanced from 3.2 to 4.4, and brand advocacy rose to 62%, with regression analyses showing strong effects (β = 0.38, Adj. R² = 0.78). Despite these implications, future research must discuss long-term behavioural validation, ethical suggestions, cross-cultural applicability, and dynamic optimization of stochasticity over various service environments to sustain meaningful engagement.
Governance, regulation, and the future of work in the age of artificial intelligence
Monisha J1
Artificial Intelligence (AI) is reshaping economies, workplaces, and societies by driving innovation, improving productivity, and creating new opportunities for growth. However, the rapid adoption of AI also raises significant concerns regarding human rights, privacy, fairness, accountability, and the future of employment. This paper examines the challenge of balancing technological innovation with the protection of fundamental human rights in an increasingly AI-driven world. The study explores the role of governments, regulatory institutions, and organizations in developing governance frameworks that promote responsible AI while ensuring transparency, inclusivity, and respect for human dignity. It reviews emerging regulatory approaches, including risk-based AI governance models, data protection regulations, and ethical AI guidelines implemented across different jurisdictions. Particular attention is given to the impact of AI on labour markets, changing skill requirements, workforce transformation, and workers' rights. The paper argues that human-centred AI governance is essential to address issues such as algorithmic bias, workplace surveillance, job displacement, and unequal access to technological benefits. It further emphasizes the need for collaboration among policymakers, industry leaders, educational institutions, and civil society to build sustainable and equitable AI ecosystems. The study concludes that effective AI regulation should not hinder innovation but rather strengthen public trust, encourage responsible technological development, and support long-term social and economic progress. Integrating human rights principles into AI governance and employment policies can help ensure that technological advancement contributes to inclusive growth, social justice, and improved quality of life for present and future generations.
Navigating accountability, law, and labour in an AI-driven society
Nisha Wadhawan1; Swati Mathur2
Artificial Intelligence (AI) systems increasingly participate in decisions that carry legal, economic, and civic weight; from hiring and credit scoring to policing and public administration. As these systems take on roles once reserved for human judgment, questions of accountability, rights protection, and labour impact have become central to responsible AI governance. This paper examines the evolving regulatory landscape surrounding AI, focusing on the tension between innovation and oversight, the "responsibility gap" created by autonomous decision-making, and the consequences of algorithmic governance for workers' rights and data privacy. Drawing on comparative analysis of the European Union's Artificial Intelligence Act, the United States' sectoral approach, and emerging international frameworks, the paper identifies structural challenges to enforcement, including regulatory lag, cross-border inconsistency, and the opacity of proprietary systems. It concludes that durable AI governance requires harmonized international standards, ethics-by-design practices, and sustained collaboration between regulators, technologists, and legal scholars to ensure that AI development remains accountable to the rights and livelihoods of the people it affects.
Implications for education and research
Megha Ojha1
The introduction of Generative Artificial Intelligence (AI) technologies has transformed the processes of creation, dissemination, and consumption of knowledge in education and research. AI systems can generate text, images, sound, software code, and creative content. Thus, AI has opened up new avenues for innovation, academic research, and information dissemination. However, the quick adoption of AI technology has led to complex issues with copyright law that current legislation does not address. Authorship, ownership, originality, fair use, licensing, and liability concerns are becoming significant as AI is increasingly used to create educational content for students, write books, and handle other writing tasks. This paper critically examines the copyright implications of generative AI for academics and researchers. The paper also discusses copyright and other IP infringement risks arising from AI-assisted content creation. Emerging judicial decisions, policy initiatives, and legal frameworks adopted across various jurisdictions have also been analyzed. More specifically, the study focuses on educators, researchers, students, publishers, and academic institutions' challenges in managing the balance between technological innovation and the protection of copyright. Using a doctrinal and comparative legal methodology, the paper evaluates whether today's copyright regimes are sufficient to deal with AI-driven creativity and content. The paper provides policy recommendations to enhance legal certainty, promote ethical AI use, and ensure sustainable coexistence between copyright protection and technological advancement in the digital age.
Drivers, multidimensional impacts, and policy implications
Viktor Radun
The growing integration of Artificial Intelligence (AI) into economic and social structures has triggered a global wave of resistance that is increasingly shaping the direction of technological change. This paper offers a multidisciplinary examination of the origins and consequences of AI resistance, drawing on economic, ethical, environmental, cultural and social aspects. Key sources of opposition identified include labor displacement, data privacy violations, ecological costs, cultural frictions, ethical issues, lack of institutional support and improper or insufficient regulations. This study frames the AI resistance as a sophisticated adaptive response to perceived threats and unmet societal expectations. The paper proposes a global AI resistance ecosystem model linking resistance triggers to their economic, social, ethical, and environmental repercussions, while also exploring policy strategies that reconcile innovation with responsible oversight. By interpreting resistance as a legitimate and informative force within democratic governance, this study argues that the central question is not whether resistance will shape AI, but how its insights can be constructively integrated into more efficient, strategic, sustainable, ethical, pro-human and socially accountable AI use in future, to the benefit of all community.
Jesubukade Emmanuel Ajakaye1; Jide Ayoola2
The rapid advancement of artificial intelligence (AI), particularly generative and agentic AI systems, is transforming work practices, professional roles, and employment patterns across multiple sectors. While existing studies have extensively examined the implications of AI for job displacement, augmentation, and workforce restructuring, considerably less attention has been paid to whether higher education institutions are adapting sufficiently to prepare graduates for an increasingly AI-driven labour market. This gap raises important questions regarding the alignment between university curricula, emerging workplace demands, and graduate employability. This paper synthesises current evidence on the relationship between AI-driven labour market transformation and higher education preparedness. Drawing on 32 peer-reviewed studies identified through a PRISMA-guided search process, the review examines how AI is reshaping occupational tasks, the skills and competencies increasingly valued by employers, and the extent to which universities are responding through curriculum redesign, pedagogical innovation, and employability initiatives. Particular attention is given to the transition from occupation-specific training towards the development of transferable capabilities such as AI literacy, critical thinking, adaptability, problem-solving, and human-AI collaboration. The findings reveals a structural misalignment between the pace of AI-driven workplace transformation and the speed of institutional adaptation in higher education. Drawing on evidence from the reviewed literature, the study presents a five-stage conceptual framework linking AI advancement, workplace transformation, evolving skill requirements, university adaptation, and graduate employability outcomes. The framework provides a structured lens for assessing institutional readiness in the age of AI and identifying areas of misalignment between educational provision and labour market needs. The paper concludes with recommendations for universities, policymakers, and industry stakeholders seeking to prepare future graduates for evolving professional environments.
A multiple-case narrative inquiry into AI-assisted work, creativity, ethical boundaries, and educational implications
Tri Pujiani1; Zohaib Hassan Sain2; Cashandra Candice Mara3
Generative artificial intelligence (GenAI) is reshaping how people complete professional, creative, and language-mediated work. While AI is often celebrated for productivity, less is known about how users maintain human judgment, originality, quality, and ethical responsibility when AI becomes part of everyday practice. The purpose of this study was to investigate the ways in which AI is used in different fields, how users manage the use of AI technology in their creativity and ethics, and what implications of such experience can be drawn for education, including language education. The qualitative multiple-case narrative inquiry method was used, involving five participants working in the field of academic activity, entrepreneurship, creation of an e-book, professional translation, social media content production, and book writing for educational purposes. The data were collected via semi-structured interviews and analyzed using thematic narrative analysis. The results reveal that AI served as a partner in productivity, a producer of the first drafts, a partner in thinking, a language improver, and as a facilitator of the workflow. However, AI output was not accepted by participants without any criticism. Users revised, checked, contextualized, and humanized the output of AI in order to retain originality, precision, professionalism, and ethical responsibility. The conclusion of the study is that the use of AI technology is connected with competences, as well as linguistic, critical, creative, and ethical judgment. It implies that education should prepare learners with AI literacy, prompt literacy, critical reading, editing, fact-checking, translation awareness, digital communication, multimodal content creation, and ethical authorship so future generations can use AI responsibly in academic, professional, and creative contexts.
Selmi Djihane1
The integration of artificial intelligence technologies into educational systems has introduced profound governance challenges that demand careful ethical scrutiny and institutional preparedness. Algeria presents a particularly instructive case , a country navigating the pressures of digital modernization in education while working to align its national development ambitions with the growing imperatives of responsible AI deployment at the academic and pedagogical levels. This paper examines the state of AI governance within the Algerian educational context through a critical theoretical lens, analyzing the evolving relationship between the country's national digitalization strategy in education and the pressing need for a coherent, rights-based framework to guide AI integration across learning institutions. Drawing on institutional theory, the study explores how Algeria is positioned to develop regulatory and ethical approaches that reflect its specific educational realities, linguistic diversity, and pedagogical traditions. The paper argues that meaningful AI governance in Algerian education requires building locally grounded ethical frameworks that center student and teacher accountability, safeguard cultural and linguistic identity in learning environments, and ensure equitable access to AI-driven educational tools. This case enriches broader scholarly debates on AI governance across the Global South, offering the Algerian educational context as a meaningful lens through which universal ethical principles can be thoughtfully adapted to diverse national realities positioning Algeria as an active and sovereign contributor to global conversations on ethical AI in education .
A comparative analysis of BRICS and EU regulatory approaches
Sudeshna Sarkar1
This study examines how divergent artificial intelligence governance frameworks in BRICS nations and the European Union reflect competing economic interests and explores their implications for international trade, investment flows, and multilateral AI governance. The EU has established a comprehensive, rights-based, risk-stratified regulatory model centered on the AI Act, emphasizing trustworthiness, fundamental rights, and supranational coordination . In contrast, BRICS countries exhibit fragmented national approaches with limited group-level policy harmonization, prioritizing digital sovereignty, data localization, and state-led development strategies. This regulatory divergence is driven by distinct economic development models, geopolitical positioning, and competing visions of technological autonomy.The analysis reveals critical gaps. BRICS is predominantly treated as a juridical or political grouping rather than an economic policy bloc, and rigorous macroeconomic impact studies comparing regulatory regimes are absent. The paper synthesizes governance-focused and international relations studies to map the contours of this regulatory fragmentation, identify its economic drivers and consequences, and propose pathways toward more coherent multilateral AI governance that balances innovation, sovereignty, and equitable development.
Artificial Intelligence as a Catalyst for Economic Connectivity Between the Diaspora and the Home Country
Miodrag Simić1; Radovan Milovankić2; Marko Pavlović3
Contemporary developments in the digital economy increasingly highlight the importance of the diaspora as a valuable economic, business, and communication resource for countries of origin. However, existing mechanisms for connecting diaspora communities with domestic markets often remain fragmented, insufficiently personalized, and largely confined to traditional institutional or promotional channels. Within this context, digital marketing and artificial intelligence (AI) create opportunities for developing hybrid business models that integrate digital platforms, data analytics, automated communication, and human networks of trust.
This paper aims to examine the role of AI-supported hybrid business models in digital marketing as a mechanism for strengthening economic connectivity between diaspora communities and their countries of origin. Particular attention is devoted to exploring how artificial intelligence can facilitate diaspora segmentation, personalized communication, predictive identification of business interests, and more effective engagement among market participants, investors, and institutional stakeholders. The anticipated findings suggest that AI can serve as a catalyst for transforming existing approaches to diaspora engagement, particularly through more precise audience targeting, automated customer support, behavioral analytics, the development of personalized offerings, and the establishment of more sustainable digital business ecosystems. Hybrid business models enable the integration of technological efficiency with the social capital embedded within diaspora networks, thereby enhancing opportunities for investment, exports, tourism, knowledge transfer, philanthropy, and the development of entrepreneurial initiatives. The significance of this study lies in its conceptualization of the diaspora not merely as a demographic or institutional phenomenon but as an active economic segment that can be strategically engaged through digital marketing and AI-driven systems. The implications of the study are relevant to policymakers, chambers of commerce, marketing agencies, investment networking platforms, and businesses seeking to develop more effective strategies for communication, economic engagement, and collaboration with diaspora communities.
Governance challenges in the AI economy
Kamel Dhay1
This research examines the new frontier of entrepreneurship and digital work by Syrians in the context of economic reopening, post-war recovery, and increasing global digital integration. As the world undergoes labour-market transformation driven by artificial intelligence (AI), online labour platforms, and digital business tools, new opportunities are opening up for Syrians to earn money, become self-employed, and start enterprises. At the same time, significant governance challenges remain regarding regulation, infrastructure, labour protections, access to finance, digital inclusion, and ethical use of AI systems. The study asks whether these technologies can reduce traditional barriers to entrepreneurship, including geographical isolation, limited formal employment, and restricted market access, or whether they risk reproducing inequality through algorithmic bias, weak institutional oversight, and unequal digital capabilities. Using a qualitative exploratory approach, the paper draws on 6 semi-structured interviews with Syrian entrepreneurs. This analysis is theoretically grounded in Mixed Embeddedness Theory, which explains entrepreneurship as shaped by individual skills, digital tools, and institutions, showing Syrian digital entrepreneurship as structurally embedded within evolving platforms, AI systems, and uneven governance contexts. The findings, interpreted through Mixed Embeddedness Theory, imply that the Syrian entrepreneurial opportunities developing through the AI economy are co-determined by personal resources (including skills, digital literacy, and networks) and emerging opportunity structures (including AI-facilitated platforms and global digital labour markets) and limited institutional environments (characterised by unequal regulation, infrastructural constraints, and fragmented governance). The benefits of AI-driven digital ecosystems can be seen as potential sources of income and self-employment, but their long-term positive effects depend on the institutional conditions that actively channel those benefits.
Regulation, rights, and the future of work in the age of AI
Ridhima Gupta1
Artificial Intelligence (AI) has emerged as one of the most transformative technological developments of the twenty-first century, reshaping education, employment, public administration, healthcare, communication, economic production, and social relations. The rapid expansion of generative artificial intelligence has intensified debates regarding the appropriate relationship between technological innovation, democratic governance, fundamental rights, and the future of work. Although AI has the potential to improve productivity, expand access to knowledge, support personalised education, assist scientific discovery, and create new forms of employment, its deployment also produces significant risks relating to privacy, discrimination, surveillance, misinformation, labour displacement, algorithmic opacity, intellectual property, and unequal access to technological benefits. These challenges demonstrate that AI governance cannot be treated merely as a technical or commercial issue. It is fundamentally a question of law, rights, social justice, institutional accountability, and democratic control. This paper examines the governance of artificial intelligence through a systematic and thematic review of existing academic literature, international policy frameworks, regulatory developments, and institutional reports. The study focuses primarily on four interconnected dimensions: governance and regulation; fundamental rights and social justice; education and knowledge; and work and labour transformation. The analysis demonstrates that existing AI governance approaches are increasingly moving from voluntary ethical principles towards more structured, risk-based and rights-oriented regulatory frameworks. However, significant gaps remain in enforcement, global coordination, accountability, transparency, and protection for vulnerable communities. The European Union's risk-based regulatory approach, UNESCO's human-centred guidance on generative AI in education and research, the NIST AI Risk Management Framework, and the United Nations' emerging proposals for global AI governance illustrate the diversity of contemporary regulatory responses. The paper argues that effective AI governance requires a multi-level and human-centred framework based on accountability, transparency, privacy, non-discrimination, human oversight, labour protection, democratic participation, and equitable access. In particular, AI regulation must move beyond the narrow objective of preventing technological harm and address the structural distribution of power, wealth, data, and decision-making authority. The future of AI governance should therefore be understood not as a choice between innovation and regulation, but as a process of ensuring that innovation remains compatible with human dignity, fundamental rights, social equality, democratic institutions, and decent work.
Rania Ahmad1; Subrato Kumar Dey1
Despite the significant transformation that artificial intelligence has brought about in the paths and mechanisms of developing many different businesses and economic activities, knowing the type, size, and directions of transformation that artificial intelligence brings about in labor markets in the context of small and medium enterprises adopting this knowledge revolution still needs more study and analysis. This study explores the reality and impact of the use of artificial intelligence (AI) in a sample of small and medium enterprises (SMEs) in the Syrian Arab Republic, through a set of indicators that illustrate the transformation in the number and type of workers, as well as the transformation in job descriptions and required skills, in addition to exploring the social responsibility of these enterprise owners in relation to their role in qualifying and training workers and raising their capabilities to reduce their dismissal or replacement with new workers . The results of the analysis of data collected using an e-questionnaire indicate that the adoption of AI in Syrian SMEs may still be in its early stages, but some indicators may serve as an early warning of an upcoming transformation, with the expectation of increased adoption in the future, which calls for proactive policies and plans to support the positive transformation in labor markets. The paper presents a set of recommendations that may help decision-makers and experts to develop policies that focus on the development of small and medium enterprises in particular and link them to educational policies that aim to achieve consistency and harmony between the development of educational outputs and the transformation in labor markets brought about by artificial intelligence in these markets.
Lessons from Nigeria's 2018-2023 electoral reforms for Africa
Njoku Angelina Chigozie1
This study examines 'Re-engineering electoral systems for electoral democracy, lessons from Nigeria's 2018 to 2023 electoral reforms for Africa'. The specific objectives of the study are to identify legal and technological shortcomings in Nigeria's elections from 2018 to 2023, to examine how these shortcomings affected public trust in Nigeria's elections, and to draw key lessons for re-engineering electoral democracy in Africa. The issue is that public trust in Nigeria's 2023 election was brought low by poor laws and inadequate technology. Using secondary data from reports, statutes, and journal articles, the study employed a qualitative design. The Technology Acceptance Model and Digital Governance Framework served as a framework for the thematic analysis of the data. The results reveal two primary gaps: first, the Electoral Act 2022 was unclear about uploading IReV results in real-time. Second, IReV failed due to inadequate networks and training. BVAS was successful in voter checks. Trust in INEC dropped to 23% as a result of these flaws. The study concluded that technology cannot foster trust on its own. Early testing, trained personnel, and clear legislation must all cooperate. The most important lessons for Africa are to test all electoral technologies before using them, train personnel early, and adopt clear rules.
Challenges in attributing liability in light of comparative legal frameworks
Omeir Hadjira1
The rapid development of artificia intelligence (AI) technologies has opened up significant opportunities for economic growth, the promotion of digital transformation and the support of innovation in Algeria. Conversely, the expanding use of artificial intelligence systems in automated processes and decision-making activities has contributed to the emergence of new forms of criminal risks and unlawful behaviour. These developments have raised fundamental legal issues concerning the determination of criminal liability where offences are committed by or with the aid of artificial intelligence systems. This study aims to examine criminal liability for AI-related offences under Algerian law, by analysing the adequacy of the provisions of the Algerian Penal Code in addressing offences involving AI systems, given that traditional rules of criminal liability are based primarily on human intent, personal fault and the principle of individual liability. The study also addresses the legal issues arising from the intrinsic and evolving characteristics of advanced artificial intelligence systems, particularly with regard to determining liability amongst programmers, developers, manufacturers, operators and users. The study adopted an analytical legal approach to assess the suitability of current criminal law rules in addressing emerging crimes linked to artificial intelligence, It also highlighted the legislative challenges posed by technological developments and emphasised the need to adapt the legal framework to ensure the establishment of effective criminal liability, whilst at the same time continuing to encourage technological innovation and support the digital transformation process. The study concludes that the existing Algerian legal framework remains insufficient to address all the legal issues arising from crimes linked to artificial intelligence, which necessitates legislative reforms aimed at clarifying the rules for attributing criminal liability, enhancing legal certainty, and providing effective protection against the risks arising from increasingly autonomous and complex artificial intelligence systems.
Implications for education, labour, and society
Rutvi Meena1; Asit Balwantrai Acharya2
Artificial Intelligence (AI) is no longer just a technological innovation; it has become a powerful force reshaping governance system, human rights frameworks, and the future of work across societies. Its growing use in education, labour markets, and public administration offers significant opportunities for efficiency, personalization, and innovation. At the same time, it raises serious concerns regarding accountability, transparency, fairness, and the protection of human dignity. This paper examines the evolving landscape of AI governance, focusing on how regulatory frameworks attempt to balance rapid technological advancement with ethical responsibility. It highlights key challenges such as algorithmic bias, risks to data privacy, and the need to safeguard digital human rights. Addressing these challenges requires strong oversight mechanisms and inclusive, participatory policymaking that reflects diverse social perspectives. In the context of education, the integration of AI into universities is transforming teaching, learning, and assessment practices, demanding new governance approaches to ensure fairness and academic integrity. In the labour sector, the paper explores the socio-economic impact of automation, the expansion of the gig economy, and the urgent need for reskilling and upskilling to prevent widening inequalities. Adopting an interdisciplinary perspective, the study situates AI within broader social and cultural contexts, drawing on insights from law, ethics, sociology, and technology studies. It argues for a model of responsible AI governance that prioritizes human well-being, protects labour rights, and promotes inclusive and equitable development. Ultimately, the paper contends that effective governance is essential to ensure that AI serves as a tool for social progress rather than a source of exclusion and disparity.
Samet Kurtağzı1
Since the emergence of Bitcoin in 2009, crypto assets have proliferated globally as a decentralised and cross-border hybrid of technology and finance, transforming the traditional financial intermediation model. Türkiye stands out as one of the leading countries worldwide in crypto asset usage, owing to its young and techsavvy population, high inflation environment, and exchange rate volatility. For a long time, Turkish law lacked a clear definition and comprehensive regulatory framework for crypto assets; under Law No. 6493, only their use as a payment instrument was prohibited, while trading, custody and taxation remained largely unregulated. This situation changed radically with Law No. 7518 on the Amendment of the Capital Markets Law, which entered into force on 2 July 2024, bringing crypto asset service providers under the licensing, supervision and sanctioning authority of the Capital Markets Board (CMB). Adopting a descriptiveanalytical method and a comparative law perspective, this study examines the current state of crypto asset law and tax law in Türkiye in light of Law No. 7518, the CMB's Communiqués III35/B.1 and III35/B.2, the December 2024 amendments to the MASAK regulations, and the crypto asset transaction tax and withholding tax regime adopted in early 2026. By comparing Türkiye's regulatory model with those of Japan, the United States, Switzerland and the European Union (MiCA), the study finds that the current framework has achieved significant maturity in supervision and licensing, while still containing gaps in tax legislation, investor protection mechanisms and judicial precedent. The conclusion offers concrete recommendations for legal practitioners and policymakers.
Cicih Ratnasih1; Syaiful1; Mohammad Rizal Tamami1; Eka Tofa Kumalasari1
This study aims to analyze the effect of training and technology usage ability on employee performance, and its impact on increasing compensation for Contact Center agents at Bank Permata, Jakarta. Targeted training is expected to enhance work competencies, while technological mastery is a key requirement to support the effectiveness of call center services. Both factors are believed to contribute to achieving more optimal performance, which in turn affects increased compensation as a form of company recognition of employees.
Insights from diverse professional sectors
Vesna Radojcic1; Pavle Dakic2; Sasa Ljubojevic3; Nebojsa Todorovic4
The growth of artificial intelligence (AI) and automation is a major factor changing the global labor market. It greatly affects job structures and the shift in traditional roles. This paper looks at how AI systems are causing certain jobs to disappear while also pointing out new types of work that need specific skills. In addition to reviewing relevant literature, we surveyed 40 people from various educational and professional backgrounds. The goal was to understand their views on how AI affects the labor market, what skills are now needed, and the importance of ongoing professional development. Most respondents think that AI will greatly change current jobs, but it will also create new roles. They stress the importance of further education and developing digital skills. Both the literature review and survey results highlight the need for ongoing education, reskilling, and careful management of technology changes to maintain human talent. The study concludes that we should not see AI just as a threat to jobs. Instead, with the right policies, planning, and human oversight, it can help foster long-term economic and social growth.
Nur Çelik Efşan1
It is known that technological developments have influenced interstate relations from the past to the present. While the development of transportation routes formed the basis of diplomatic relations in traditional diplomacy, in the 21st century, with the development of studies on artificial intelligence, digitalization activities in the private sector have spread to the public sector, and artificial intelligence dominance is observed in a wide range of areas in diplomacy, from intelligence gathering and increased security measures to electronic service delivery and the preparation of international political and commercial texts using special software. All of these have paved the way for the digitalization of diplomacy. Indeed, digital diplomacy is a foreign policy tool through which a country maintains its connections with other states through negotiations and mutual cooperation without resorting to hard power. The announcement of digital diplomacy to the public through the new media order has increased transparency, and with the removal of communication barriers due to the impact of globalization and the widespread use of internet technologies in state administration, it has taken its place as a tool for influencing and informing individuals. Digital diplomacy; While facilitating the conduct of diplomatic relations online, it is also a supporting factor that enables states to maintain a strong image in the international arena. The aim and importance of this study is to analyze the advancement of diplomatic relations on the basis of artificial intelligence in our century shaped by artificial intelligence. A literature review was conducted in the study
Ethical challenges and social responsibility in the transformation of work, culture, and human relations
Nikhil Thakur1; Syed Asghar Mehdi2; Shilpi Bhatia3
Background: The hospitality industry has seen the rise of artificial intelligence adoption and at the same time, bleisure tourism has become a widely popular segment of travel. These two trends are changing the way services are delivered, however, they are also raising ethical issues such as privacy, employment, and human interaction. Objective: This paper is aimed at investigating the ethical dilemmas and social responsibility aspects of the use of artificial intelligence in bleisure tourism and hospitality, especially in terms of changes in work, culture, and human interaction. Methodology: A systematic review of the literature was carried out using the databases Scopus, Web of Science, Emerald Insight, ScienceDirect, and Google Scholar. The focus was on peer-reviewed articles published from 2014 to 2025. A thematic content analysis was performed. Results: AI is capable of significantly improving the efficacy of operations and the level of personalization for bleisure tourists. Nevertheless, the issues of data security, discrimination by algorithms, loss of jobs for employees, monitoring of employees, and the diminishing of genuine human contact have all been highlighted as problems. Conclusion: In order to incorporate artificial intelligence into bleisure tourism and hospitality in a way that is sustainable, it is crucial to find a point of equilibrium between technological advancements on the one hand and the ethics of responsibility, the values of human-centered service, and social sustainability over the long term on the other.
AI, labour rights, and professional ethics
Ashok K. Rathoure1
The inclusion of artificial intelligence in schools is quickly changing how we teach, manage, and support students. While AI offers improved efficiency, personalization, and access, it also raises significant issues regarding workers' rights, professional independence, and ethical practices in education. This paper looks closely at the growing conflicts between AI-driven automation and the human-centered values that have long defined educational work. Drawing on insights from labour law, educational ethics, and technology studies, the analysis looks at three main aspects of AI's effect on educational work. First, it examines the loss of teacher autonomy due to management by algorithms and surveillance systems. Second, it discusses how AI could reduce the skills needed by educators while shifting professional responsibilities to automated platforms. Third, it highlights the failure of current legal frameworks to deal with issues like liability, consent, and collective bargaining rights in AI-driven educational settings. This article argues that without active governance, AI could worsen unstable working conditions, lower professional judgment, and weaken the important relationships in teaching. On the other hand, if AI is developed with input from educators and based on principles of rights, it can enhance professional practice instead of replacing it. The article ends with practical suggestions for policy-makers, institutional leaders, and teacher unions to ensure that the future of work in education is driven by ethical awareness, social discussions, and a strong dedication to labour rights.
Strengthening corporate ethical accountability in educational technology enterprises
Alok Kumar1
Artificial Intelligence (AI) has become the backbone of the educational‑technology (EdTech) ecosystem, transforming how content is delivered, assessed, and tailored to learners. Yet, algorithmic opacity, data bias, and the ecological costs of intensive computation raise ethical and sustainability concerns. Investors and regulators increasingly demand transparent Environmental, Social and Governance (ESG) reporting that accounts not only for carbon and labor metrics but for digital integrity. Drawing on stakeholder and stewardship theories, this research investigates how AI‑governance maturity influences ESG‑reporting quality and corporate ethical accountability in global EdTech enterprises. Employing a mixed‑methods design (quantitative survey n = 60 firms; qualitative interviews n = 15), regression and mediation models reveal a significant positive effect of AI‑governance mechanisms - ethics boards, algorithmic audits, and bias‑control training—on ESG disclosure comprehensiveness (β = 0.54, p < 0.001). Ethical accountability partially mediates this relationship, transforming compliance into stakeholder trust. Theoretically, the paper contributes a construct of Algorithmic Accountability as Governance Capital (AAGC) linking digital responsibility with sustainability performance. Managerially, it proposes integrating AI metrics within ESG dashboards, dual financial/AI audit certifications, and sector‑specific reporting guidelines. These findings advance understanding of responsible innovation and position EdTechs as pioneers of transparent, ethics‑centric digital sustainability.
The Relationship with Perceived Decision-Making Autonomy and Responsibility
Bojana Nešić1
This study examines how the use of and reliance on artificial intelligence in employee selection processes affect HR professionals’ and candidates’ perceptions of decision-making autonomy and responsibility. The aim is to investigate how greater use of AI changes individuals’ sense of personal control and responsibility in decision-making, as well as to examine whether these effects differ between the two groups. Existing research has focused primarily on the efficiency, acceptance, and ethical frameworks surrounding the use of artificial intelligence in HR processes. However, the psychological consequences of AI use, particularly in terms of perceived decision-making autonomy and responsibility, remain insufficiently explored. Moreover, differences in these outcomes between HR professionals and candidates have received relatively little research attention. The study will employ a survey questionnaire to assess the use of and reliance on artificial intelligence and perceptions of control over decision-making. The constructs are based on the Trust in Automation Scale (Jian et al., 2000; Körber, 2019) and perceived behavioral control (Ajzen, 1991). It is hypothesised that greater use of and reliance on artificial intelligence will be associated with a reduced sense of control over decision-making and a partial shift of responsibility from the individual to the technological system. These effects are expected to differ between HR professionals and candidates, contributing to a better understanding of behavioural patterns associated with the use of artificial intelligence in employee selection.
“When Free Isn’t Free”: Biometric Signatures and Mechanisms of Manipulation in the Digital Ecosystem
Aleksandra J Dimitrijević1
Contemporary digital marketing has moved beyond traditional methods of user tracking. This review paper synthesises insights from digital communication studies, cognitive framing, and data science. It focuses on the collection of metadata that is not readily apparent to the average user, not through direct audiovisual surveillance but through subtle forms of sensor-generated data. The paper examines the role of accelerometers, gyroscopes, and users’ biometric signatures in the creation of detailed psychological profiles. Drawing on Shoshana Zuboff’s theoretical framework of surveillance capitalism, it explores how algorithms use physiological parameters derived from users’ interactions with their devices to predict emotional states. The mechanisms underlying personalised advertising can be compared to aggressive telephone sales techniques, but may achieve greater success because they allow for automated targeting of individuals’ cognitive vulnerabilities. The ultimate aim extends beyond selling products. The continuous measurement of user behaviour through the Android system provides data for deep learning, whereby the user ceases to be merely a customer and becomes a source of raw material for the development of artificial intelligence. The paper highlights the ethical challenges of an era in which privacy is exchanged for the appearance of free access to digital services, emphasising the need for critical digital literacy.
Generation x's perceptions of SGFinDex in Singapore's retirement planning ecosystem
Daniel Tan1
Singapore's demographic landscape is increasingly defined by the dual pressures of an accelerating ageing population and a critically low fertility rate, necessitating a strategic shift toward individualised retirement adequacy. Within the "Smart Nation" framework, the Singapore Financial Data Exchange (SGFinDex) was established as the world's first foundational public digital infrastructure to resolve data fragmentation for the "sandwich generation" (Generation X). This qualitative study examines the perceptions of this cohort (born 1965–1980) through three semi-structured focus groups and reflexive thematic analysis. The findings reveal five core themes: Visibility Without Guidance identifies a structural lacuna between data consolidation and the "algorithmic production of meaning" necessary for actionable retirement insights; Trust as Context suggests that while high institutional trust serves as a baseline, it remains decoupled from the perceived security of downstream data usage by private entities; Usability Friction highlights barriers in re-linking requirements and mobile interface habits; Fragmentation Anxiety illustrates how partial data integration creates a significant "cognitive burden"; and Advisor Mediation positions human intermediaries as essential "adoption multipliers." These results expand the Technology Acceptance Model (TAM) by critiquing "pure" digital-only governance models. As the MyMoneySense application is decommissioned, this research advocates for the integration of Human-AI hybrid models and Explainable AI (XAI) to provide the interpretive guidance currently missing. By addressing the intersection of social responsibility and digital governance, the study offers a critical perspective on how public trust in technological solutions is maintained through mediated interaction within complex socio-technical systems.
Junaid Kc1; Divya Nv2
AI is increasingly becoming part of the education and business value chain, and this is changing the scope and nature of entrepreneurship training and competency building. While traditional entrepreneurship education has focused on creativity, innovation, problem-solving, leadership, and opportunity recognition, the advent of AI-powered technologies is reshaping the methods and implications of teaching, learning, and practice in modern digital economies. This paper critically analyses the contribution of AI in revolutionising the definition of entrepreneurial skill development in entrepreneurship training and management education. This study examines how AI-powered business planning tools, intelligent mentoring systems, predictive analytics, simulation platforms, and generative AI applications are utilised in entrepreneurial learning settings. It also examines the role of AI technologies in training models to foster digital adaptability, strategic thinking, innovation capacity, data literacy, and decision-making skills among emerging entrepreneurs. Although AI holds promise for customised learning, improved market analysis, and faster idea generation, it also introduces new ethical and technological concerns, such as the risk of a dependence on AI, a decline in human creativity, and unequal access to AI tools. Based on current interdisciplinary literature, the paper emphasises the importance of integrating AI complementary, human-centric, and ethically responsible within entrepreneurship training systems. The research adds to the academic debate on the direction of entrepreneurship education and highlights the need to foster adaptive and resilient entrepreneurial skills in the era of AI.
A review of evidence from developing economies
Anjan Niyogi1; Pallavi Kumari2
The rise of Artificial Intelligence in higher education is transforming learning, student assessment, academic management, and institutional strategies in diverse global contexts. In this paper, the transformative impact of AI-driven teaching technologies on enriching personalised learning, adaptive assessment, predictive analytics, and evidence-based decision-making at universities. Motivated by the Technology Acceptance Model and institutional change theory, this study investigates the role of digital readiness, faculty acceptance, policy orientation, and socio-cultural background that will affect AI adoption in academic ecosystems (Davis, 1989; Zawacki-Richter et al., 2019). Using a comparative analysis of emerging scholarship and policy environments, the paper examines institutional experiences across Canada, the European Union, the Balkan Peninsula, MENA, and representative countries of the Global South. The comparative findings suggest that universities in Canada and the European Union are well equipped with AI governance frameworks, as well as having strong digital platforms, regulatory environments, and strategic investment. Meanwhile, organizations in the Balkans, MENA, and the Global South are innovative, even as they grapple with infrastructure, ethical, and policy obstacles. The study identifies, among other things, continuing concerns about algorithmic bias, academic integrity, data privacy, and the maintenance of pedagogical autonomy (Selwyn, 2019). Sustainable institutional reform, it contends, requires human-centered, ethically governed, and contextually respecting AI-embedded implementation approaches that can ensure a well-rounded balanced approach between technological advancement and educational equity, cultural diversity, and long-term academic resilience.
Cheryl Cheng Xin Yi1
Generative artificial intelligence (GenAI) has emerged as a rapidly developing technology with significant potential to enhance the effectiveness and quality of managerial decision-making. However, there are many potential concerns associated with the impact of GenAI outputs on managerial trust. These include the lack of transparency, data bias, errors, inaccuracies, and inconsistencies. This study aims to investigate the evolution of managerial trust in GenAI outputs and its impact on decision-making across the financial institutions. It draws on theories from Trust-in-automation, technology adoption, human–AI collaboration and organizational governance. This study then constructs a conceptual framework and empirically tests the relationships among the GenAI adoption, evaluation capabilities, internal control, and verification, and trust in GenAI outputs. Meanwhile, it explores the mediating role of trust between GenAI adoption and managerial perception of effective decision-making. An online survey was distributed to obtain the primary data for this study. A total of 100 valid responses were obtained and analyzed using SmartPLS4 to examine the structural relationships among the constructs. The results revealed that frequent usage has a significant positive impact on trust towards GenAI outputs. It led to higher confidence among managers in the decisions made. Furthermore, trust in GenAI outputs acts as a partial mediator between usage and decision-making effectiveness. The findings suggest that trust is formed more directly through experiential use of GenAI than through governance structures. The study identified actions that organizations can take to ensure ethical practice of GenAI-assisted decision-making.
José G. Vargas-Hernández1; Patricia Villasana-Ramos2; Omar C. Vargas-González3
The intersection of artificial intelligence (AI) and green finance represents one of the most consequential developments in Mexico's transition toward a low-carbon economy. This research report examines how AI-driven technologies—including machine learning, natural language processing, deep neural networks, and predictive analytics—are reshaping the landscape of renewable energy finance in Mexico. The findings contribute to an emerging body of knowledge that positions Mexico as a middle-income country navigating both the opportunities and the structural constraints associated with AI-enabled green finance. Policy recommendations emphasize the need for open data ecosystems, AI governance frameworks aligned with national energy transition goals, and public-private investment in digital literacy and climate fintech capacity.
Opportunities, ethical challenges, and future implications
Shivansu Sachan1
Artificial Intelligence (AI) is transforming the global hospitality industry by redefining customer interaction, service personalization, operational efficiency, and guest satisfaction. This paper examines the role of AI-driven technologies such as chatbots, virtual assistants, predictive analytics, facial recognition systems, and smart hotel management platforms in enhancing guest experiences and building long-term customer loyalty. The study further explores how AI contributes to personalized services, faster response systems, automated decision-making, and improved service quality in hotels and tourism establishments. At the same time, the paper critically analyzes the ethical and social challenges associated with AI adoption in hospitality, including concerns related to data privacy, surveillance, employment displacement, algorithmic bias, and reduced human interaction. Through interdisciplinary perspectives combining tourism studies, management, technology, and ethics, the research highlights the need for responsible and human-centered AI implementation. The paper also discusses the future implications of AI for hospitality education, workforce development, and sustainable tourism practices. It argues that while AI can significantly improve operational effectiveness and customer satisfaction, maintaining emotional intelligence, cultural sensitivity, and human values remains essential in hospitality services. The study concludes that a balanced integration of technological innovation and human interaction is necessary for creating ethical, inclusive, and sustainable guest experiences in the age of artificial intelligence.
A comparative study on Singapore's regulatory sandbox with the Indian regulatory mechanism
Nirankush Kenjige1; Chaitra Rangappa Beerannavar2
A 'regulatory sandbox' typically involves testing of new products or services in a controlled, regulatory environment. This helps to gather evidence on the benefits and risks of a new financial innovation, while monitoring and managing the risks. The sandbox provides a structured way for regulators and the innovators, to interact with the ecosystem, and assist the regulators to develop regulations that encourage innovation and enable the delivery of relevant, low-cost financial products. It is an important tool that can create a more dynamic, evidence-based regulatory environment that helps to adapt the various emerging technologies. The paper devolves around the niche 'regulatory sandbox' mechanism as followed by Singapore, alongside a comparative analytical study with jurisdiction of India, and India's stance on the regulatory mechanism. The paper aims at analysing the regulatory fallacies in the Indian system by analysing the risks involved, and thereafter looking at the security and financial implications. Ever since the increased valuation and societal consideration of the block chain technology and the secured nodal transactions, the impact and importance it gained, in order to surface a regulatory sandbox in Singapore - being a notable and imminent participant in the technological advancements attached to the financial technology- is inevitable. From P2P lending to the perils of Insurtech, Singapore has been evolving towards the creation of a regulatory sandbox for the fintech industry and hence the understanding of the same plays a pivotal role in the overall Performa. Start-ups such as ADDX and BondEvolve are paving the way to provision of direct access to the financial market and thereby, assisting in the general mitigation in Singapore. Hence, the progress, and risk factors attached with it, require in-depth detailing and introspection
Nagendra Singh Yadav1; Vishal Kumar Goar2; Pallavi Singh Yadav3
This paper examines how the introduction of LLMs (Large Language Models) into commercial software is driving a structural transformation of work, management practices and economic value capture across the software industry, a transformation with consequences for professions, institutional adaptation and digital labour markets. For nearly two decades, SaaS (Software as a service) business models rested on a single load-bearing assumption that the marginal cost of serving an additional user was effectively zero. This assumption justified flat rate subscriptions, i.e. all you can eat pricing tiers and the 75-85 percent gross margins that defined cloud software economics through the 2010s. The integration of LLMs (Large Language Models) into commercial software products has broken that assumption. Every model inference now consumes real, metered compute and the cost scales with prompt length, response length, model size and request frequency. We document this structural shift, develop a quantitative framework, the inference Cost Propagation Framework for understanding how token-level prices set by foundation model providers cascade through application layer business models and analyze the strategic responses now visible in industry data. Drawing on May 2026 pricing from OpenAI and Anthropic, public company disclosures and the ICONIQ Capital 2026 State of AI survey, we show that average AI product gross margins have settled near 52 percent against a SaaS benchmark of 75-85 percent, with inference alone consuming roughly 23 percent of revenue at scaling stage firms. The paper closes by identifying which firms, professions and institutional arrangements are most exposed to these new economics and what governance and management responses are emerging
Poornima Mahesh1
This paper proposes a rights-based regulatory design to address harms caused by AI systems in the workplace, centering remedies and redress mechanisms that protect workers' autonomy, privacy, and livelihoods. centring remedies and redress mechanisms that protect worker autonomy, privacy, and livelihood. The Objectives are threefold: (1) map the main categories of AI-related workplace harms (surveillance, automated decision-making in hiring, task allocation and performance evaluation, and opaque disciplinary algorithms); (2) assess existing legal and institutional remedies across jurisdictions with emphasis on gaps in applicability; and (3) propose a workable regulatory architecture that integrates statutory rights, accessible remedies, and organizational accountability measures. to map the principal categories of AI-related workplace harms (surveillance, automated decision-making in hiring, task allocation and performance evaluation, and opaque disciplinary algorithms); (2) to evaluate existing legal and institutional remedies across jurisdictions with emphasis on enforceability gaps; and (3) to propose a practicable regulatory architecture that integrates statutory rights, accessible redress pathways, and organizational accountability measures. The Methodology combines doctrinal legal analysis, comparative policy review (India, EU and select Global South contexts) and qualitative case synthesis of recent litigation and regulatory actions involving platform and corporate employers. EU, and selected Global South contexts), and qualitative case synthesis of recent litigation and regulatory actions involving platform and enterprise employers. The analysis highlights recurring obstacles – burden of proof, opacity of evidence, regulatory fragmentation and power asymmetries between workers and employers – and examines tools such as algorithmic impact assessments, mandatory explainability standards, collective bargaining supplements and specialized ombudsman or court processes. obstacles—burdens of proof, evidentiary opacity, regulatory fragmentation, and power asymmetries between workers and employers—and examines tools such as algorithmic impact assessments, mandated explainability standards, collective bargaining complements, and specialized ombuds or tribunal processes. Expected contributions include a modular regulatory model that links the articulation of rights to procedural remedies, the design of principles for enforceable remedies (including presumptions of transparency and evolving burdens of proof), and policy recommendations tailored to jurisdictions with limited enforcement capacity. articulation to procedural remedies, design principles for enforceable redress (including presumptions of transparency and shifting evidentiary burdens), and policy recommendations tailored for jurisdictions with limited enforcement capacity. The paper is aimed at policymakers, labor regulators and academics who are looking for concrete frameworks to ensure that the use of AI at work respects fundamental rights and provides rapid and effective remedies. targets policymakers, labour regulators, and scholars seeking actionable frameworks to ensure AI use at work respects fundamental rights and provides timely, effective remedies.
Felipa Lopes Dos Reis1; Bernardo1; Afonso1; Rui Castaneira1
This research sets out to explore and compare how quality management is perceived and implemented in two Worten retail locations — specifically, the store in Rio Sul and the store in Colombo. The main goal is to analyze the sales sector through the eyes of the customer, particularly focusing on how feedback provided by consumers can influence the adaptation or improvement of a particular quality dimension within these environments. To carry out this comparative analysis effectively, the research adopts a methodology based on the SERVQUAL model — a widely recognized tool in the domain of quality management. This model facilitates the evaluation of service quality by examining the gap between customers' expectations and their actual perceptions of the service received. By working through this gap, researchers can understand not only how well a service is being delivered but also how closely it aligns with what customers value. The SERVQUAL framework comprises five main dimensions that provide a comprehensive view of service quality: tangibility (which relates to the physical appearance of facilities, equipment, and staff), reliability (referring to the ability to perform the promised service accurately and dependably), responsiveness (which captures how willing and prompt employees are in providing service), assurance (the knowledge and courtesy of staff, as well as their ability to instill trust), and empathy (the extent to which personalized attention and care are offered to customers). A structured questionnaire, grounded in these five dimensions, was designed and distributed to customers at both Worten stores. The responses gathered from this questionnaire form the empirical base of the study and are analyzed to identify strengths and weaknesses in service quality at both locations. By interpreting these data, the research aims to offer a detailed and data-informed comparison, shedding light on how service quality is experienced differently in two branches of the same retail chain. Ultimately, this study seeks not only to assess the current state of service quality management at the Rio Sul and Colombo Worten stores but also to identify actionable recommendations that could enhance customer satisfaction. The research contributes valuable insights for management teams looking to refine their approach and tailor their services more effectively to customer expectations, particularly in a highly competitive and dynamic retail sector.
Tarana Afrin Chandel1
Conventional teaching methodologies are rapidly being replaced by digital-tech education with flexible and personalized learning. These learning options have opened the doors to micro-credentials, enhancing the employability rates of students. Universities and higher education institutes (HEIs) have started taking this into consideration. Aiming to enhance employability by strengthening industry-academia collaboration, this study investigates how these two parameters affect learner outcomes and global employability. Covering the period from 1990 to 2025, the study analyzes how significant skills gaps, conventional practices, and static higher education degrees hinder long-term career success. This study examines the influence of AI-related micro-credentials on the employment designations for students in higher education. A quantitative correlational design was utilized to examine the relationships among the AI-Driven Skill Model (AIS), Job Market Transformation (JMT), Micro-Credentialing System (MCS), AI-Driven Curriculum Redesign (ACR), and Global Employability (GE). Data indicates a substantial positive correlation between the acquisition of micro-credentials and employability. The data demonstrates that micro-credentials serve as clear indicators of job readiness, linking educational programs with workforce requirements. This highlights the significance of incorporating micro-credential programs into bachelor's curricula to improve graduate job prospects in the digital age.
Revisiting the j-curve hypothesis
Riddhi Mehta1; Arshad Bhat2
The J-curve hypothesis has been used in long time to analyze the relationship between the exchange rate fluctuations and adjustment in trade balance especially in developing economies. This paper re-tests the existence of the bilateral J-curve effect in the case of India in monthly data and in a multivariate time-series model that both accounts for the short-run dynamics and the long-run relationships. The evaluation is based on the bilateral trade of India, with the 5 top trading partners, namely the United States, China, Japan, Germany, and the United Arab Emirates during the years 1991-2024. Unit-root tests, Johansen cointegration tests and Vector Error Correction Models are utilized where they are statistically and diagnostically appropriate. The findings suggest that bilateral trade relations of India do not depict a strong/systematic J-curve effect. There is also long-run cointegration of the limited number of trading partners and no long-run consistent relationship between the real exchange rate and the trade balance of China and Japan. Even though cointegration is found to be the case in Germany, the instability of the parameters is indeed high, due to probable macroeconomic and structural shocks within the Eurozone, which restricts the accuracy of the dynamic VECM estimation. In the case of the United Arab Emirates and the United States, there are weak and statistically insignificant adjustment dynamics at the error-correction mechanisms and impulse response analyses after the exchange rate shocks. Altogether, the results indicate that the adjustment mechanisms, based on exchange rates, have a minor impact on the bilateral balance of the Indian trade, and the structural factors prevail in the dynamics of trade. These findings have significant implications to the exchange rate management and other external sector policy in India.
Emerging challenges in ownership, creativity, and patent governance
Varanasi Rahu1
Generative Artificial Intelligence (Generative AI) has revolutionised creativity and innovation across fields such as education, business, media, healthcare, and the tech industry, transforming how people think and work. Today, AI tools like ChatGPT, Gemini, Midjourney, and DALL·E can create content independently from humans in all forms: text, images, software code, designs, and inventive solutions, which now pose a threat to traditional notions of authorship, ownership, inventorship, or legal responsibility. The current IPR systems are mainly designed to safeguard works of human creativity, but with the introduction of AI-generated works, significant legal, ethical, and governance issues have arisen. This study explores the changing landscape of the relationship between Generative Artificial Intelligence and Intellectual Property Rights, with a special focus on ownership, creativity, and patentability, as well as patent governance in the digital innovation economy. The paper uses a conceptual research approach and an analytical approach based on doctrinal legal analysis, comparative policy evaluation, and literature review of current literature, international legal cases, and new AI governance frameworks. It examines issues such as copyright conflicts with AI-generated works, claims to inventorship, ethical issues with data training, limits on current patent laws, and algorithmic accountability. The study also makes a valuable contribution to interdisciplinary discussion by incorporating law and management, innovation studies, and technology governance perspectives, and by offering conceptual suggestions for developing sustainable and responsible innovation ecosystems that leverage AI technologies.
Vikas Rana1; Prakash Meena2; Ravneet Singh2
The Dharamshala region of Himachal Pradesh has become an important cultural tourism destination because of its rich Himachali traditions, Tibetan influence, heritage temples, local cuisine, handicrafts, folk music, spiritual environment, and natural landscape. Cultural experiences significantly shape how tourists perceive a destination, their emotional connection, and their overall satisfaction during their visit. At the same time, the rapid rise of Artificial Intelligence (AI) in tourism services has changed how tourists get information, plan trips, enjoy experiences, and give feedback. AI technologies, including personalized recommendation systems, chatbots, virtual guides, sentiment analysis, and smart tourism platforms, increasingly affect tourist expectations and satisfaction. This study aims to explore how Himachali cultural experiences impact tourist satisfaction in the Dharamshala region and to look into the role of AI-enabled tourism services in improving cultural engagement and visitor experiences. The research will focus on cultural aspects like local festivals, traditional food, heritage attractions, folk art, spiritual practices, local hospitality, and the authenticity of the destination. It will also examine the factors that affect tourist satisfaction, such as service quality, cultural interaction, accessibility, digital tourism support, safety, destination image, and AI-based travel assistance. The study will use a quantitative research design based on primary data collected from 220 domestic and international tourists visiting Dharamshala, McLeod Ganj, and nearby tourism spots. A structured questionnaire using a five-point Likert scale will gather data through convenience and purposive sampling techniques. Statistical tools, including Descriptive Analysis, Reliability Test (Cronbach's Alpha), Exploratory Factor Analysis (EFA), Correlation Analysis, and Multiple Regression Analysis, will analyze the relationship between cultural experiences, AI-based tourism services, and tourist satisfaction. The study's findings are expected to emphasize the growing importance of combining AI technologies with regional cultural tourism to improve tourist satisfaction and promote sustainable destination development. This research will contribute to smart tourism, cultural heritage management, and AI-driven hospitality practices while offering practical insights for tourism stakeholders and policymakers in Himachal Pradesh.
Dana-Marie Ramjit1
The governance of artificial intelligence presents one of the most consequential institutional challenges of the contemporary era, yet prevailing policy responses remain anchored in hierarchical, state-centric models of that are structurally misaligned with the distributed, non-linear, and polycentric character of AI systems themselves. This paper argues that heterarchical governance offers more analytically precise and practically relevant tool for understanding this misalignment that currently dominates AI governance discussions. Drawing on the author's prior work theorizing heterarchical models of digital governance and administrative legitimacy (Ramjit, 2025), the paper develops the concept of digital heterarchy as an original analytical framework for understanding how algorithmic authority operates across and between formal institutional hierarchies, generating accountability discrepancies and legitimacy crises that monocentric models are structurally ill-equipped to resolve. The paper advances three interconnected arguments. First, heterarchy reveals why AI systems resist hierarchical governance, their authority is distributed, contested, and operates simultaneously across multiple scales, rendering centralized oversight persistently reactive and institutionally inadequate. Second, existing AI governance arrangements, including the EU AI Act and emergent national regulatory regimes, reproduce the very hierarchical logic that digital heterarchy exposes as insufficient, creating governance gaps with significant consequences for democratic accountability and public trust. Third, the paper proposes that legitimate and socially responsible AI governance requires a relational, multi-scalar, and citizen-centered institutional design, rooted in heterarchical principles rather than hierarchical command. The paper concludes with comparative implications for institutional design across the EU, Canada, the Balkans, and the MENA region, with particular attention to the practical relevance of heterarchy theory for policymakers, regulators, and scholars navigating an era of contested algorithmic authority.
Pedro Pinheiro1; José Luis Silva2
Artificial intelligence has become a strategic priority for governments, universities and firms, but its responsible adoption depends not only on technological capacity but also on institutional governance. This paper examines national readiness for AI governance using open international data from the International Monetary Fund's AI Preparedness Index, complemented, where relevant, by policy information from the OECD.AI Policy Observatory and the Oxford Insights Government AI Readiness Index. Focusing on the relationship between digital infrastructure, human capital, innovation capacity and regulation and ethics, the study analyses how countries differ in their ability to support responsible and trustworthy AI. The empirical analysis compares selected European countries with different levels of economic development, digital maturity and regulatory capacity. Particular attention is given to the governance dimension, including ethical regulation, institutional coordination, accountability and the capacity to manage risks related to privacy, bias, transparency and labour market disruption. The analysis shows that AI readiness is uneven and multidimensional. Countries with stronger innovation systems and digital infrastructure do not automatically present equivalent strengths in ethical governance, while countries with emerging AI ecosystems may face particular challenges in translating policy ambition into institutional capacity. The paper contributes to interdisciplinary debates on AI in society by shifting attention from adoption to governance. It argues that responsible AI requires more than technical investment: it depends on regulatory preparedness, public sector capacity, skills development and mechanisms for social accountability. The findings are relevant for policymakers, educators and institutional leaders seeking to align AI development with ethical principles, social responsibility and inclusive digital transformation.
Perspectives for artificial intelligence-based financial analysis
Asma Hajji1
This study examines the joint influence of environmental factors and U.S. financial markets on the returns of Bitcoin (BTC) and Ethereum (ETH). The research investigates how environmental indicators, including CO₂ emissions, environmental innovation, and ESG performance, interact with major U.S. financial market indices to affect cryptocurrency returns. The analysis incorporates key financial indicators such as the S&P 500, NASDAQ, Dow Jones Industrial Average, gold prices, and oil prices, together with environmental variables, in order to provide a comprehensive understanding of the determinants of cryptocurrency performance. The study is situated within the broader context of technological developments in financial analysis, particularly the growing use of artificial intelligence in processing large datasets and supporting financial forecasting. While the research relies on conventional quantitative and econometric methods, its findings provide useful insights that may contribute to future AI-based financial analysis and prediction models in cryptocurrency markets. By combining environmental and financial perspectives, this research contributes to the growing literature on sustainable finance and digital assets. The study seeks to provide a deeper understanding of the factors shaping cryptocurrency returns and to highlight the potential value of advanced analytical technologies in studying the evolving dynamics of financial markets.
Mohamad Ali AlHussam1
This paper challenges the conventional paradigm of crisis management by demonstrating that global economic and political turbulence should not be treated as an exogenous shock for managers to navigate, but rather as a systemic condition influenced by corporate decisions. Investigating how sustainability education prepares current (executive MBA) and future (undergraduate) business leaders for global volatility, the study presents four key insights. First, both cohorts recognize that traditional global supply chain practices accelerate global turbulence, whereas sustainable practices serve a mitigative function. Second, exposing students to supply chain sustainability benchmarks actively develops global awareness and managerial empathy. Third, empirical findings indicate that gender-balanced environments optimize sustainability-focused strategic decision-making. Finally, the research reveals that threshold learning can occur independently through structured self-reflection, bypassing the strict requirement for direct instructor facilitation. Ultimately, this study establishes that integrating sustainability into management curricula empowers leaders to understand how their supply chain choices directly amplify or alleviate macroeconomic instability.
Svetlana Veljanovska1
The rapid advancement of generative artificial intelligence (AI) is transforming higher education by offering new opportunities for teaching, learning, assessment, and research. AI-powered tools are increasingly used by students and academic staff to support content creation, information retrieval, data analysis, and academic writing. While these technologies can enhance efficiency and accessibility, their growing adoption raises important legal and ethical concerns. This paper examines the key legal and ethical challenges associated with the use of generative AI in higher education. Particular attention is given to issues of academic integrity, authorship, intellectual property rights, data protection, transparency, and accountability. The study further explores the emerging European regulatory framework governing artificial intelligence and its implications for higher education institutions. In this context, the paper analyzes the need for institutional policies and ethical guidelines that can ensure the responsible use of AI while preserving fundamental academic values. Special consideration is given to the opportunities and challenges faced by higher education institutions in North Macedonia in aligning their practices with European standards and regulatory developments. The paper argues that the effective integration of generative AI requires a balanced approach that combines technological innovation with legal compliance, ethical responsibility, and academic integrity. Such an approach is essential for fostering trust, promoting quality education, and ensuring the sustainable use of AI technologies in higher education.
J V P Udaya Deepika1
The integration of the Internet of Things (IoT) presents valuable opportunities for small businesses to strengthen their brand identity through data-driven insights. This chapter mainly focuses on the application of Time Series Forecasting techniques to analyse the temporal patterns within data generated by IoT devices deployed in small business environments. By examining trends in customer interactions, operational metrics, and product usage captured through IoT sensors, time series forecasting can reveal crucial insights into customer behaviour, predict future demand, and optimise service delivery. This research investigates the potential of algorithms such as ARIMA, Exponential Smoothing, and Prophet to extract meaningful patterns from IoT data streams, enabling small businesses to personalise their branding efforts, anticipate customer needs, and ultimately foster stronger brand loyalty. The findings aim to provide a practical understanding of how small businesses can effectively leverage time series forecasting on IoT data for impactful brand development.
A management perspective on ethics and social responsibility
Ridhima Gupta1
The rapid integration of Artificial Intelligence (AI) into organizational operations has created significant opportunities for innovation while raising important ethical and social concerns. Responsible AI adoption has therefore emerged as a critical management issue, requiring organizations to ensure that AI systems are developed and used in ways that are fair, transparent, accountable, and aligned with societal values. This paper explores Responsible AI from a managerial perspective, focusing on the strategies and practices that support ethical decision-making and corporate social responsibility in AI-driven environments. It discusses the role of leadership in establishing governance structures, promoting ethical awareness, managing potential risks, and fostering stakeholder trust. The study also examines challenges associated with AI implementation, including bias, privacy protection, explainability, and regulatory compliance. By synthesizing contemporary research and management insights, the paper highlights the need for a holistic approach that combines technological capabilities with ethical oversight and organizational responsibility. The analysis suggests that organizations that proactively embed responsible AI principles into their strategies are better positioned to achieve sustainable growth, strengthen stakeholder confidence, and contribute positively to society. The study concludes that responsible AI adoption is not solely a technical requirement but a strategic management responsibility essential for long-term organizational success.
Undergraduate students' encounters with dark patterns
Bodhisattwa Bardhan Choudhury1
Dark patterns, or misleading user interface designs that discreetly control consumer behaviour against their own interests, are becoming more common in India as digital commerce grows. There is an urgent need to evaluate how common consumers identify, encounter, and react to such activities in light of the Central Consumer Protection Authority's (CCPA) November 2023 Guidelines, which listed thirteen categories of dark patterns under the Consumer Protection Act, 2019. This study looks on undergraduate students' experiences with dark patterns in Indian e-commerce and digital platforms. Undergraduate students are a digitally active but potentially vulnerable consumer demographic. A structured online questionnaire with three sections was used to gather data from 184 undergraduate respondents using a cross-sectional survey design. These sections included demographic profiling, an experiential rating of specific dark pattern encounters (such as false urgency, basket sneaking, drip pricing, subscription traps, and disguised advertisements), and an assessment of attitudes regarding ethics, trust, and regulatory necessity. A five-point Likert scale was used to evaluate responses, and descriptive statistics and inferential tests including chi-square, t-tests, and ANOVA were used for analysis. Findings reveal the frequency and impact of dark patterns on user trust and decision-making, and highlight a significant gap between digital confidence and actual awareness of manipulative design. The study underscores the need for consumer education and stronger regulatory enforcement to safeguard digital consumer rights among India's youth.
A comparative evaluation of SARIMA, LSTM, XGBoost and MLP models
Babafemi Daniel Ogunbona1; Anas Ibrahim1
Reliable inflation forecasting is central to monetary policy formulation and macroeconomic management, particularly in structurally volatile economies. This study comparatively evaluates four forecasting models - Seasonal Autoregressive Integrated Moving Average (SARIMA), Long Short-Term Memory (LSTM) networks, Extreme Gradient Boosting (XGBoost) and Multilayer Perceptron (MLP) - applied to monthly headline inflation rate of Nigeria from January 2009 to November 2024 (191 observations). An 80:20 temporal train-test split was applied uniformly across all models. Stationarity was established after second-order differencing and seasonal analysis confirmed an annual cycle at lag 12. MLP achieved the best forecasting performance across all evaluation metrics (RMSE = 1.02, MAE = 0.86, MAPE = 3.62%, R² = 0.975), outperforming other models. XGBoost performed poorly on the test set (RMSE = 8.94, R² = -0.892) despite strong cross-validation performance (CV RMSE = 1.20), a divergence attributed to the structural inflation break of 2022–2025 that exceeded the model's training distribution range. These findings establish MLP as the most accurate standalone model for Nigeria's inflation series and demonstrate that architectural simplicity, when appropriately calibrated, can outperform more complex deep learning architectures on small macroeconomic datasets.
Amal Kouri1; Ghofrane Barhoumi1; Jamel-Eddine Gharbi1
This research examines the factors underlying the low level of economic complexity in Tunisia and, as a result, the decline in the diversification of its exports. Based on complexity theory, we studied the influence of business sophistication, market efficiency, and productive human capabilities. The data used come from the global competitiveness, global innovation and global prosperity reports. The results show that export diversification can only be reversed through a profound transformation of capabilities, business sophistication and market efficiency.
Chayma Mehouachi1; Jamel-Eddine Gharbi2
The objective of our study is to verify the extent to which the adoption of Artificial Intelligence by companies influences their marketing capabilities and performance. Data collection was carried out by sampling. Following the verification of the dimensionality, reliability and validity of the measurement scales, we checked our structure model. The results show a significant positive impact of the adoption of Artificial Intelligence on performance and a significant impact mediated by marketing capabilities.
Sibangiso Ngwenya1; Thembekile Blessmore Ndebele2; Samkeliso Suku Dube2; Caroline Mhlanga2; Admire Bhuru2; Sinqobile Caren Dube2
The rapid growth of investment in artificial intelligence (AI) has created a highly uneven global funding landscape, with capital concentrated largely in a small number of high-income economies. Despite increasing interest in AI's potential contribution to development, comparatively little is known about how AI investment is distributed across the Global South. This paper presents a secondary data analysis of funding disparities within AI startup ecosystems in Sub-Saharan Africa, Latin America and Southeast Asia, benchmarked against investment patterns in OECD member states, which are used here as a proxy for high-income innovation economies. Data were drawn from publicly available venture capital datasets, harmonised to United States dollars and deduplicated across sources. The findings reveal significant inequalities in AI investment across regions. As of mid-2025, 159 AI startups in Sub-Saharan Africa had collectively raised US$803.2 million in external funding between 2020 and mid-2025. The analysis also highlights disparities in funding concentration, sectoral focus and investment stage progression. In Sub-Saharan Africa, more than 55% of AI startup activity is concentrated in fintech, talent technology and education, while investment in firms developing foundational AI technologies remains limited. Similar patterns of application-layer concentration and pre-seed stage dominance are evident in Latin America and Southeast Asia. Drawing on decolonial AI theory, this paper argues that these disparities reflect broader patterns of structural exclusion within the global innovation economy. The study concludes by outlining policy recommendations to strengthen Global South participation in AI development through targeted investment and improved access to AI infrastructure.
Emerging skills and employability challenges for future graduates
Esra Tracey1
This study examines the growing influence of artificial intelligence on employment patterns, professional roles, and graduate readiness. It explores how AI-driven automation is reshaping the nature of work by transforming routine tasks, redefining job expectations, and creating demand for new forms of human expertise. The study identifies occupations most exposed to technological disruption while also highlighting emerging career opportunities linked to data analysis, AI literacy, digital ethics, human-machine collaboration, creativity, and complex problem-solving. It also discusses the widening gap between traditional university preparation and the evolving skills required in an AI-driven economy. Particular attention is given to the role of higher education institutions in redesigning curricula, strengthening practical learning, and helping students develop adaptable, future-oriented competencies. Ultimately, the study argues that graduate employability in the age of AI depends not only on technical knowledge, but also on lifelong learning, ethical awareness, and the ability to work intelligently alongside emerging technologies.
Opportunities, risks and governance challenges
Victor Olusoji1; Nkemdilim Iheanachor2
The possibilities of using AI to change the way the public sector works and improve the services it provides are a phenomenon that continues to be investigated by governments globally (Wirtz et al., 2019). AI offers transformative opportunities in Nigeria, a country with long-standing governance issues of inefficiency and bureaucracy that are stifling development. This paper presents an extensive analysis, which will include policy document analysis, as well as a case study analysis of pilot projects for AI in agencies in Nigeria. The findings shall shed more light on how AI can deliver measurable gains, from automatically processing citizen applications to predicting resource allocation and fraud in public expenditure. But we project that critical risks may arise: algorithmic bias can perpetuate discrimination norms; opaque AI decision-making can diminish accountability and citizen trust; public AI systems are potentially vulnerable to data breaches and cyberattacks; and AI usage can create a greater digital divide if it is not accompanied by an investment in digital literacy (Sun & Medaglia, 2019). The outcome of this paper calls for the establishment of a complete governance system, including ethical guidelines, algorithmic impact assessments, public procurement standards for AI and institutional capacity development in the field of regulation and oversight (Floridi et al., 2018). The study aims to add to the comparative AI governance literature by analysing the Nigerian context and how it differs from emerging international governance frameworks, thereby shedding light on the unique challenges of developing democratic AI governance systems.
Legal and ethical challenges of AI governance
Racha Zeghmar1
Artificial intelligence (AI) is rapidly transforming societies, workplaces, and decision making processes, creating new opportunities while also raising complex questions about governance, regulation, rights, and accountability. This paper explores the challenges of legal responsibility and liability in the age of AI, focusing on the difficulty of determining who should be held accountable when AI systems cause harm, produce biased outcomes, or operate in unpredictable ways. It examines how traditional notions of responsibility are being challenged by increasingly autonomous technologies and highlights the need for effective regulatory frameworks that balance innovation with the protection of human rights. The paper argues that responsible AI development requires transparent, ethical, and human centered governance involving policymakers, legal institutions, technology developers, and society. Addressing AI liability is not only a legal issue but also a broader challenge that will influence the future of work and human responsibility in an increasingly automated world.
India and the architecture of a Global South AI model
Mitrajit Biswas1
As artificial intelligence (AI) rapidly develops, standard global frameworks remain expensive and concentrated within a few Western tech monopolies, often leaving developing nations behind. This paper examines India's vital role in architecting a distinct "Global South Model of AI" that prioritizes affordability, inclusivity, and public welfare. Rather than acting as a passive consumer, India leverages its proven Digital Public Infrastructure (DPI) philosophy to build an accessible, public-good technology ecosystem. Central to this vision is the state-backed IndiaAI Mission, which expands sovereign computing capacity, alongside the "Trusted AI Commons"—an open-source repository of datasets and language models designed specifically for resource-constrained countries. This structural shift was highlighted at the India–AI Impact Summit in New Delhi, positioning India as a strategic bridge for technical cooperation across Africa and Southeast Asia. Built on the principles of equitable progress, India's model challenges commercial monopolies by championing open science, algorithmic transparency, and energy-efficient AI systems tailored for low-resource environments. Furthermore, India's regulatory frameworks, including the Digital Personal Data Protection Act, offer a unique mechanism for ethical data governance outside traditional US-China rivalries. Ultimately, by shifting the global narrative from corporate profits to scalable public utility, India provides the Global South with a reproducible blueprint for digital sovereignty. This approach ensures that emerging economies can leverage AI for sustainable socio-economic development without sacrificing their technological independence.
Joudy Zayat1
Artificial intelligence is becoming a powerful tool in pharmaceutical supply planning by helping factories predict raw material demand more accurately. Instead of relying only on manual estimates, AI can analyze past production data, seasonal trends, supplier performance, and market changes to forecast what materials will be needed and when . This helps reduce shortages that can stop production and overstocking that can waste money and storage space .In a medicine factory, better supply planning is important because delays or missing ingredients can affect the whole production line. AI can support planners by detecting risks early, adjusting inventory levels, and improving order timing for raw materials . As a result, the factory can work more efficiently, lower costs, and maintain a more stable supply of medicines for patients.
The place of critical thinking, judgement and imagination in the age of artificial intelligence(AI)
Rodgers Otieno Odhiambo1
The emergence of Artificial Intelligence (AI) in the global stage, has dramatically and significantly transformed not only how knowledge is created, but also accessed, and assessed. This change is especially noticeable in higher education, where research, instruction and learning are increasingly incorporating AI-powered tools like ChatGPT and large language models (LLMs). This paper examines the impact of Artificial Intelligence (AI) on legal education and practice in Kenya, highlighting its influence on critical thinking and human judgment. The paper discusses the transformative effects of Artificial Intelligence, such as increased access to information and the automation of tasks, while raising concerns of about its limitations in replicating essential human abilities like critical and evaluative judgement. The paper calls for a re-evaluation of traditional pedagogical frameworks in this new technological landscape. While AI tools like ChatGPT offer significant benefits such as automating research and improving efficiency, they also pose serious risks to traditional legal practice and education. The legal field ultimately depends on ethical judgment, nuanced reasoning, and interpretive analysis. AI cannot replicate these distinctively human faculties. As rapid AI integration continues, it is critical that academics and legal practitioners preserve and foster fundamental human skills, particularly evaluative judgment and critical thinking, within traditional pedagogical frameworks. This is because critical thinking in legal education is often associated with skill-based activities, including analyzing, synthesizing, judging critically, and evaluating problems.
Aditya Singh1; Cristina Raluca Gh. Popescu2
The development of economies, public services, in addition to governance is being transformed by AI in the recent years, which allows the immediate requirement for ethical frameworks along with regulatory policies which are effective ones. The AI Act helps European counties to have founded a thorough regulatory method which is on the basis of risk, whereas the Asian countries have implemented diverse strategies of governance which indicates distinct technological ecosystems, economic priorities, as well as legal systems. This study will explore the existing gaps in the literature, and it will evaluate principles which are ethical ones, regulation of AI, as well as governance methods in a comparative manner across both above mentioned continents via present evidence. The paper will examine the main dimensions which will comprise of human oversight, transparency, regulatory enforcement, accountability, privacy, in addition to fairness. This study will detect the continuing governance hurdles, regional differences and similarities which affects the execution of responsible AI in such countries. Then, the paper will talk about for large scale cooperation at the international level in the governance of human centered AI which is still advancing & adaptive in nature, when considering the potential opportunities, in order to balance the concerned ethical responsibility along with novelty.
Cristina Raluca Gh. Popescu1; Aditya Singh2
The digital transformation, intelligent decision making, in addition to automation, are all leading towards the reshaping of workplaces by AI in the recent years, which not only generates novel opportunities but also raises worries about the fairness, rights of the workers, governance in regard to employment, as well as algorithmic accountability. The past few years showed regulatory progress especially in European countries, which presents increasing awareness to guarding the workers employed in places where the management of the workplace is assisted by AI and its linked risks, while the hurdles on the path of its incorporation still exist. This study will perform an analysis especially a comparative one in the context of workplace modernization led by AI in not only developing but also developed countries via several available evidence, after exploring the existing gaps in the literature. The paper will explore the ways in which distinct regulatory as well as economic contexts affects the implementation of Artificial Intelligence, ethical governance, protection of employees, adaptation of organizations, in addition to reskilling of workforce. The paper will detect various hurdles associated with human oversight, bias, privacy, as well as transparency, when discussing policy methods to encourage comprehensive in addition to responsible transformation of workplaces. The paper will aid in setting up recommendations which are of practical nature to balance social responsibility when proceeding with technological novelty along with rights of workers, to assist sustainable implementation of AI on a large scale comprising of different countries.
A case study of a primary school sustainable-oriented activity
Charis Kalogera1; Lamprini Seremeti1
The integration of Environmental, Social and Governance (ESG) principles into education represents a paradigm shift in preparing educational stakeholders to design sustainability-oriented activities. This study discusses the motivation behind the adoption of ESG principles in K-12 schools, the categories of educational activities in which K-12 schools engaged, and the institutional mechanisms that affect their implementation through the prism of ESG. The aim is to highlight how K-12 schools adapt to global norms and expectations concerning sustainable development. To this purpose, the study focuses on a specific educational activity that has been implemented in a primary school. The activity concerns the investigation of the history of traditional foods within the school's local territory, with the aim of connecting students with local cultural heritage, collective memory and community identity. Through interviews with older residents, the collection of traditional recipes, the study of local agricultural products and food reservation practices, students investigated how food reflects the history, geography, economy and social life of the region. From an ESG perspective, this activity is described by using the Social, Environmental and Governance dimensions, as well as the institutional mechanisms that affect its implementation under each dimension. By examining this school-based activity through the ESG lens, the study demonstrates how sustainability can be assessed not only through quantitative indicators but also through qualitative evidence derived from everyday educational practices.
An integrative review
Ruchika Shokeen1; Nazli2; Rishu Raj3
Background: Artificial Intelligence (AI) is increasingly transforming educational environments through personalized learning systems, intelligent tutoring platforms, assistive technologies, and data-driven educational practices. For learners with intellectual disabilities, AI offers significant opportunities to improve accessibility, communication, participation, and individualized learning experiences. However, the use of AI also raises ethical concerns related to autonomy, informed consent, privacy, algorithmic bias, digital exclusion, and human dignity. Despite the growing discourse on AI ethics, the specific needs and rights of persons with intellectual disabilities remain underrepresented in existing literature. Methodology: This conceptual paper adopts an integrative literature review methodology. Relevant literature published between 2020 and 2026 was identified through databases including Google Scholar, ERIC, Scopus, and Web of Science using keywords related to artificial intelligence, intellectual disability, inclusive education, ethics, accessibility, and social responsibility. In addition, key policy documents such as the UNCRPD and UNESCO's Recommendation on the Ethics of Artificial Intelligence were reviewed. The selected literature was analyzed thematically to identify major opportunities, ethical challenges, and principles for the responsible use of AI in inclusive education. Results: The review identifies both opportunities and challenges associated with the use of AI for learners with intellectual disabilities. AI-based technologies can support personalized learning, enhance communication, promote independence, and facilitate educational participation. At the same time, concerns emerge regarding data privacy, accessibility barriers, algorithmic discrimination, reduced human agency, and unequal access to digital resources. Five key ethical dimensions were identified as central to the responsible use of AI: human dignity, autonomy, privacy, accessibility, and social responsibility. Conclusion: The paper argues that the ethical implementation of AI in inclusive education requires a disability-inclusive approach that prioritizes equity, participation, transparency, and respect for human rights. Educators, policymakers, researchers, and technology developers share responsibility for ensuring that AI technologies promote meaningful inclusion rather than reinforcing existing inequalities. Ethical and socially responsible AI has the potential to contribute significantly to educational access and social participation for persons with intellectual disabilities.
Hafsi Latifa1; Hafian Insaf2
Technical disruptions, ranging from projector failures and electricity interruptions to unstable internet connectivity, have become a challenging daily routine in many educational settings. This low-resource teaching environment severely compromises teacher motivation, instructional productivity, and student learning outcomes. This paper provides an analytical description of these systemic challenges using institutional data, highlighting the urgent need for scalable solutions to the ongoing issues hindering educational objectives. Utilizing an applied descriptive methodology, this study explores the potential of Artificial Intelligence (AI) as both a supportive instructional tool and a contingency framework to mitigate the impact of infrastructural deficits. The findings suggest that fostering teachers' digital literacy and integrating targeted AI tools can substantially enhance pedagogical performance, bridge the digital divide, and cultivate resilient learning environments despite resource constraints.
AI as a catalyst for inclusive education - universal design for learning, accessibility, and special needs support
Pratheesh Abraham1; Kripa Mathew2
The growing acceptance of Artificial Intelligence (AI) technology across educational settings offers a transformative but varied possibility to strengthen inclusive education. Despite decades of legislative frameworks and policy campaigning, including the United Nations Convention on the Rights of Persons with Disabilities (CRPD)(2006) and different acts all over the world, fundamental hurdles to equitable learning remain in everyday classrooms. The current study explores how the combination of AI- driven pedagogical tools and the Universal Design for Learning (UDL) might assist in eliminating structural exclusions in formal educational institutions. The study provides a formal evaluation of the present systematic review (2018-2026) employing the OpenAlex and Web of Science databases from Schools and universities in both the Global North and South. The study focuses on how AI applications, including adaptive learning systems, intelligent tutoring systems (ITS), computer vision for sign language, and emotionally responsive agents. Findings disclose that the AI technologies visibly enhance accessibility, especially for learners with visual impairments, dyslexia, autism spectrum conditions, and hearing disabilities. The paper concludes by proposing the Responsible AI-UDL Integration Model (RAIM) that focuses on disabled learners and educators. The study contributes to the growing scholarship on AI ethics in education that disability justice is an indispensable dimension of socially responsible AI.
Understanding human relationships in disability support systems
B.V. Ram Kumar1; Wasim Ahmad2
Background: Artificial Intelligence (AI) is increasingly integrated into disability support through assistive technologies, communication tools, healthcare applications, educational platforms, and AI-enabled companions. While existing research has largely focused on accessibility, autonomy, and ethical concerns, limited attention has been given to how AI influences the human relationships that are central to disability support. This paper explores the impact of AI on interactions among persons with disabilities, family members, caregivers, educators, healthcare professionals, and AI companions. Method: This conceptual review synthesizes contemporary literature on AI, disability studies, assistive technologies, human rights, and relational ethics. Relevant scholarly articles, reports, and policy documents were examined to identify emerging themes related to AI-mediated relationships in disability support settings. Results: The review indicates that AI can strengthen disability support by enhancing communication, promoting independence, reducing caregiver burden, facilitating social participation, and providing personalized assistance. However, several challenges were identified, including emotional dependency on AI systems, reduced human interaction, privacy concerns, trust issues, and the risk of replacing meaningful human relationships. Based on these findings, a Relational AI Framework for Disability Support is proposed, emphasizing the role of relationships as a key mechanism through which AI influences inclusion and well-being. Conclusion: AI has the potential to transform disability support not only through functional assistance but also through its influence on human relationships. The effectiveness of AI should therefore be evaluated in terms of its ability to strengthen human connection, preserve dignity, and promote inclusive participation. Human-centered and ethically grounded approaches are essential to ensure that AI complements rather than replaces supportive human relationships.
Opportunities, challenges, and ethical implications
Samah Sellami1; Jamel-Eddine Gharbi2
The rapid development of artificial intelligence is transforming education systems worldwide through the introduction of innovative tools such as adaptive learning platforms, intelligent tutoring systems, and generative AI applications. These technologies are increasingly recognized for their potential to broaden access to knowledge, personalize learning experiences, and improve educational outcomes. In this context, artificial intelligence is viewed as a promising lever for women's educational empowerment and for reducing structural inequalities in access to education. However, despite these opportunities, the integration of artificial intelligence into education raises major concerns regarding gender inequality, algorithmic bias, and unequal access to digital technologies. This article examines, through an in-depth literature review, an integrated theoretical framework, and a quantitative methodology based on PLS-SEM modeling, the extent to which artificial intelligence contributes to women's educational empowerment, while identifying the main barriers related to the digital divide and algorithmic discrimination. The study also explores the ethical dimensions of AI adoption in education, particularly issues of transparency, fairness, and inclusion, and underscores the importance of developing gender-sensitive AI systems and policies that promote digital literacy and equitable access to technology. Finally, it highlights the need for responsible governance of artificial intelligence to ensure that technological innovation aligns with the principles of social justice and equality. Overall, this research contributes to the ongoing debate on artificial intelligence and education by offering a critical perspective on its dual role as both a vector of empowerment and a potential source of inequality, particularly for women in educational contexts.
A structural equation modelling approach to educational equity and social inclusion
Parul Kulshrestha1; Rajni Bala2
Purpose: This research explores the impact of digital divide on educational equality and social inclusion in AI-based educational environments through the influence of AI literacy, participation in AI-based education, inappropriate algorithm, and information asymmetry, and inclusive AI governance. Method: The study employed a quantitative and cross-sectional approach and distributed a structured questionnaire to 400 respondents comprising students, teachers, faculty, administrators, and educational technology experts in Punjab, India. The theoretical framework used for this research was developed based on Digital Divide Theory, the Capability Approach, the Technology Acceptance Model, Algorithmic Fairness Theory, and Responsible AI Governance Theory. Data collected was analyzed using SPSS and AMOS 28.0 software through CFA (Confirmatory Factor Analysis) and CB-SEM (Covariance-Based Structural Equation Modelling). Results: Results show that digital divide adversely affects educational equality and social inclusion, while AI literacy, involvement in AI-based education, and inclusive governance are positively related to educational equality. Originality: Utilizing a blend of technology, education, and governance, this research provides a better understanding of responsible artificial intelligence (AI) use and also delivers useful advice to decision-makers in policy and educational institutions as well as technical developers who want to establish inclusive, fair, and sustainable AI education.
International students caught in the institutional crosshairs in U.S. higher education
Rafael Arsen Melkonian1
The fast adoption of generative artificial intelligence has transformed higher education, creating new opportunities for learning while simultaneously raising concerns about academic integrity and institutional control. Using the California State University system's large-scale adoption of ChatGPT as a case study, this paper examines how international students navigate the contradictory position of being encouraged to use AI tools while and questions about academic dishonesty through AI detection systems. Drawing on Medium Theory, Neoinstitutional Theory and Reception Theory, the project conceptualizes AI not simply as a tool for students, but as an increasingly important element within the academic environment that reshapes learning, changes institutional expectations, and potentially affects the self-image of internationals students when AI detection systems render false positives. As the largest public university in the U.S., with over 470,000 enrolled students, the California State University system is a bellwether for how other public universities in the U.S. may see comparable issues. This research investigates how the CSU system has simultaneously promoted AI as a symbol of innovation while deploying policies and detection technologies intended to regulate its use. The study combines analysis of academic literature and institutional discourse with qualitative interviews conducted with international students and higher education stakeholders. Preliminary findings suggest that international students primarily use AI for language support, comprehension of course materials and academic productivity rather than plagiarising. At the same time, participants report uncertainty regarding acceptable AI use that have been shown to disproportionately misclassify non-native English writing as machine-generated.
Soumya T. Varghese1; Jonah Angeline2
To ensure equitable access for students with disabilities such as visual impairments, ensuring the adaptation of inclusive education models in mainstream schools is a priority. The literature regarding inclusive practices is mainly seen among three domains such as technological, spatial, and systemic. This review is an attempt to search, map, and analyse such literature about inclusive adaptations. The search is specifically focusing on inclusive practices for students with visual impairments in mainstream classrooms. The research is contextualised using the Joanna Briggs Institute methodology for search and extraction. Then the PRISMA-ScR reporting standards are used to analyse the content from the Scopus database. The Population-Concept-Context framework has been shared with the independent reviewers for the extraction verification process. The qualitative thematic analysis was conducted to identify focus clusters, technological patterns, and critical research gaps. The review will help educators and educational policymakers to get an overview of practical difficulties encountered while implementing inclusive frameworks.
Renata Šimunović1
In the time we live in, it is impossible not to expose a child to electronic media because they are part of our everyday life. The media can equally be useful, but also harm the child's development through inadequate use. For this reason, their use should be subject to certain rules, time limits, and it is definitely necessary to pay close attention to the choice of content to which the child is exposed. The goal of this research was to examine as many people as possible, in order to get a total and complete impression of their experience related to the topic being studied. The results of the research conducted on 32 people showed that most of them (59.4%) were exposed to electronic media in childhood. Also, 53.1% do not have an opinion on the acceptability of children's exposure to electronic media. Almost 72% of respondents encountered the positive influence of the media on children, while almost 50% of them saw violence in the media as a child.
Challenges for education and social inclusion
Parul Kulshrestha1; Rajni Bala2
Learning, work and communication have been significantly transformed through Artificial Intelligence (AI) creating new possibilities for social and economic growth. However, there is an unequal access to AI resources and therefore, this unevenness will increase already existing problems with digital inequality. This interdisciplinary research looks at how digital inequality affects educational attainment and social inclusion in this new AI world. For example, Learning Management Systems (LMS), Adaptive Learning Technologies (ALT), and Intelligent Tutoring Systems (ITS) offer individualized ways to facilitate personalized learning. However, marginalized populations will not have access to digital resources, adequate training or AI knowledge making them unable to benefit from technological advances in education and social services. The examination of educational, technological, sociological and public policy perspectives shows the impact of algorithmic bias, inequity of data, exclusionary language and varying access to technology, as well as how these factors contribute to enlarging opportunity gaps and increasing social exclusion in education. Furthermore, the study addresses the ethical and socially responsible aspects of these technologies by stressing inclusive design, equitable access to AI, transparent governance, cultural responsiveness in the development of the technology, and establishing a common language across various disciplines. This study presents findings through a primary literature review and an analysis of recently developed policy frameworks and thus, demonstrates successful strategies for digital equity, such as: AI literacy training for teachers and students, creating and implementing equitable educational policy, supporting public-private partnerships to implement and promote digital equity, and creating human-centered AI development. Furthermore, the findings indicate that addressing digital inequity is necessary for establishing digital equity, so that AI will be used as a means of empowerment rather than exclusion in digital society. The conclusion of this study is a set of recommendations for creating equitable, inclusive, and socially responsible AI ecosystems that promote educational opportunity and social engagement for all members of society.
Rethinking educational inequality in the age of generative AI
Aysun Doğutaş1
Expectations that AI-supported learning can enhance access, personalization, and educational possibilities have been raised by the quick spread of generative artificial intelligence (AI) in education. But the growth of AI also raises significant concerns about whether or not students have equitable access to these tools. The AI divide is a new aspect of educational inequality that is examined in this essay. The study makes the case that variations in students' educational experiences and results cannot be explained by access to AI technologies alone, drawing on sociological viewpoints on the digital divide, digital inequality, and cultural capital. In addition to having unequal access to AI technologies, students may also differ in their digital skills, confidence, AI literacy, usage patterns, and ability to assess and interact critically with AI-generated information. Socioeconomic status, gender, language, cultural capital, and the resources offered by educational institutions all influence these disparities. As a result, the article views AI inequality as a complex phenomenon that goes beyond the conventional issue of technology access. It also takes into account the potential for generative AI to both reproduce or create new forms of inequality while simultaneously lowering some educational obstacles. The study offers a framework for comprehending the social circumstances that influence who gains from AI-supported education by tying discussions of generative AI to well-established sociological ideas of educational inequality. The importance of shifting from an access-oriented approach to a more thorough understanding of equitable AI participation and educational opportunity is emphasized in the paper's conclusion.
Sonti Zelma Mokobane1; Sibongile Dayimani2
The integration of Artificial Intelligence (AI) into education presents a significant opportunity to foster inclusive curricula that cater to the diverse needs of all learners. AI-driven tools can facilitate personalized learning experiences, addressing the unique challenges and potential of each student, thereby minimizing disparities and promoting inclusivity. This study investigated how an inclusive curriculum can be leveraged through AI in secondary schools in the Eastern Cape. This interpretative qualitative study was conducted in 2 secondary schools in Buffalo City municipality in the Eastern Cape. Purposive sampling was used to sample teachers as participants in the study. 40 teachers voluntarily participated in the study. Semi-structured interviews were conducted with teachers. Observations were also employed to observe how teachers use AI tools to enhance student engagement. All ethical issues were adhered to. Findings revealed that teachers welcomed the use of AI because AI-enabled tools can automate administrative tasks, freeing up teachers' time to focus on individual student support and curriculum development, which are crucial for creating inclusive learning environments. Teachers indicated that numerous barriers still prohibit them from successfully using modern technology in the classroom to promote inclusion. They further believed that some digital materials, such as those designed for assistive technologies, may be neither compatible nor accessible. Since AI is an essential component of education, the study recommended that the Department of Basic Education should ensure accessibility and provide adequate training for teachers to effectively integrate AI into inclusive education.
Umut Haydar Coskun1
Although inclusion has become a widely accepted policy goal in higher education and the number of students with disabilities in universities has increased considerably, these students continue to face structural barriers both in accessing higher education and in achieving full academic participation within it. This indicates that students with disabilities may remain invisible in higher education not only numerically, but also at institutional and pedagogical levels. This presentation examines this invisibility not as an individual limitation, but as a structurally produced outcome of inaccessible educational environments, unequal resource distribution, low expectations, and limited institutional responsiveness. Drawing on the social model of disability, cumulative disadvantage theory, Bourdieu's concept of capital, and the notion of pseudo-inclusion, the presentation argues that students with disabilities may be formally present in higher education while still being excluded from full academic participation. In this sense, inclusion may often remain nominal when accessible materials, assistive technologies, individualized support, assessment accommodations, and inclusive pedagogical practices are not meaningfully provided. The presentation also discusses the potential of artificial intelligence to both transform and complicate this landscape. AI-supported tools, such as automated transcription, adaptive learning platforms, generative AI-based tutoring systems, and accessibility technologies, may enhance educational participation. However, when these systems are designed around normative assumptions about body, language, cognition, speed, voice, literacy, and user behavior, they may reproduce existing inequalities in algorithmic form. The central argument of the presentation is that artificial intelligence should not be evaluated merely as an innovative educational tool, but also in terms of the extent to which it expands the real agency, accessibility, and participation of students with disabilities. The presentation concludes by emphasizing the need for disability-sensitive, accessibility-oriented, and socially responsible approaches to artificial intelligence in higher education.
Lovelesh Gupta1
This research explores the lived experiences, challenges, and coping strategies of LGBTQ+ parents raising children in a predominantly heteronormative society. As societal norms continue to evolve, non-traditional family structures face both newfound recognition and persistent discrimination. Using a pilot-tested survey and in-depth analysis of data from 300 LGBTQ+ parents across urban and semi-urban settings in India, this study investigates how parental identity, social support systems, and institutional interactions shape their parenting journeys. The study adopts a mixed-methods approach combining quantitative data with qualitative narratives, thereby offering a comprehensive insight into the multi-layered experiences of LGBTQ+ families. Key areas of focus include societal acceptance, access to healthcare and education, legal rights, and intergenerational communication. Pilot testing ensured instrument validity and guided refinements in survey structure and interview prompts.
A case study of Mali's youth and digital transformation
Daouda Toure1
The Sahel, particularly Mali, has a paradox. Its youth population (over 65% under 25) is a demographic dividend but structural barriers like limited access to quality education, high unemployment and digital inequality stop them from participating in the digital economy. Artificial intelligence is increasingly being promoted as a development tool. Without ethical frameworks, AI risks exacerbating existing inequalities. How can AI be leveraged to promote social justice for Malian youth? This study asks. This research adopts a qualitative case study approach, utilizing analysis of existing digital infrastructure and AI policies in Mali, comparative review of AI initiatives in similar fragile contexts, and interviews with Malian students and young entrepreneurs. Initial results indicate that AI literacy among Malian youth is very low. Most of the AI applications are imported without being adapted to the local linguistic realities (French, Bambara, Arabic). Lack of electricity and internet penetration (33%) limits AI deployment. However, mobile phone penetration (over 85% in urban areas) opens a door for AI-powered mobile learning. A preliminary ethical framework is proposed in the paper based on four principles: accessibility, localization, participation and accountability. Without intentional social justice frameworks, AI-driven development in Mali risks leaving the majority of youth behind while creating a new digital elite. This research advocates for South-South cooperation, especially between Mali and China, to build inclusive and locally embedded AI capacity.
Dian Damayanti1; Ninuk Lustyantie2; Hapzi Ali3; Prabowo Subianto4; Toni Toharudin5; Emmanuel Jean Michel Frédéric Macron6; Anton Nugroho7; Anter Venus8; Iwan2; Betania Kartika9; Syamsiah Badruddin10; Paisal Halim10; Muhammad Ikhsan Setiawan11; Zudan Arif Fakrulloh12; Pribadiyono13
Background: Children are among the most vulnerable populations in disaster contexts, facing not only physical risks but also profound psychological trauma that can impede long-term development, educational attainment, and social cohesion. Traditional post-disaster psychosocial support often relies on trained mental health professionals, which is resource-intensive and inaccessible in many low-resource settings. Micro-learning animations—short, visually engaging, and easily deployable digital tools—offer a scalable alternative for delivering trauma-informed, child-friendly content that promotes psychological first aid, emotional regulation, and a return to normalcy through culturally grounded narratives. Objective: This study developed and evaluated a micro-learning animation titled "Child-Friendly Culture for Trauma Rehabilitation," designed to support children's psychosocial recovery after natural disasters within the framework of the Sustainable Development Goals (SDGs), particularly SDG 3 (Good Health and Well-Being), SDG 4 (Quality Education), and SDG 11 (Sustainable Cities and Communities). The intervention aims to reduce post-traumatic stress symptoms, rebuild a sense of safety through familiar cultural symbols and routines, and empower children with basic self-regulation techniques. Methodology: The research employed the PLOM (Plan-Learn-Observe-Measure) design-based framework. In the Plan phase, we identified target user needs through consultations with child psychologists, disaster response agencies, and educators in three disaster-prone regions (Indonesia, the Philippines, and Haiti). A 4-minute motion-graphics animation was scripted, featuring locally recognizable cultural elements (e.g., traditional games, communal storytelling, simple breathing exercises) delivered by a child-friendly animated character. The Learn phase involved expert validation (n=5 child trauma specialists) and iterative refinement of content, achieving a Content Validity Index of 0.91. The Observe phase comprised a mixed-method pilot study with 60 children aged 7–12 years who had experienced a natural disaster within the preceding 12 months, recruited through community health centers. Participants were randomly assigned to either watch the animation (intervention group, n=30) or receive standard printed psychosocial materials (control group, n=30). Outcomes were measured at baseline, immediate post-intervention, and 2-week follow-up using the Child Revised Impact of Events Scale (CRIES-8) and a locally adapted Emotional Regulation Knowledge Quiz. The Measure phase calculated the Microlearning Effectiveness Index (MEI) for knowledge gain and effect sizes for trauma symptom reduction. Results: Preliminary findings from the Observe phase indicate that the intervention group demonstrated a 37% reduction in mean CRIES-8 scores from baseline to 2 week follow-up (Cohen's d = 0.72, moderate to large effect), compared to a 12% reduction in the control group. Knowledge of basic self-regulation techniques increased by 52% in the intervention group versus 18% in the control group. The MEI was calculated at 68.3%, exceeding the 60% benchmark for effective public health microlearning. Qualitative feedback from children and caregivers highlighted that the use of familiar cultural symbols (local folk characters, traditional songs) reduced anxiety and increased engagement. Children reported feeling "less scared" and more confident in using taught techniques (e.g., "breathing like a dragon") during aftershocks or disaster anniversaries. Conclusion: This study demonstrates that a culturally grounded micro-learning animation, developed through the PLOM framework, is an effective, scalable, and low-cost tool for trauma rehabilitation among disaster-affected children. The intervention aligns directly with SDG Target 3.4 (promote mental health and well-being), SDG 4.7 (education for sustainable development and global citizenship), and SDG 11.b (disaster risk reduction). Future research should expand to a multi site randomized controlled trial with longer follow-up and integrate the animation into national school based disaster preparedness curricula. The PLOM methodology offers a replicable blueprint for creating and validating peace oriented, child friendly media across diverse cultural contexts.
A design-based research approach
Ratan Sarkar1
Students with mild intellectual disabilities (MID) face persistent and multifaceted challenges in acquiring writing proficiency, stemming from cognitive, linguistic, and motor skill deficits that are inadequately addressed by conventional instructional approaches. The proliferation of artificial intelligence (AI) tools in educational settings has opened promising avenues for delivering personalized, adaptive, and non-stigmatizing writing support to this underserved population. This study employed a Design-Based Research (DBR) methodology to systematically develop, implement, and evaluate an AI-assisted writing intervention for five adolescents (aged 14–17 years, IQ 50–70) enrolled at Vivekananda Lokasiksha Niketan, a semi-rural residential educational institution in West Bengal, India. The intervention was structured as a four-module, progressively scaffolded curriculum—encompassing sentence writing, paragraph writing, functional writing, and narrative/creative writing—delivered across 16 sessions over seven weeks. Eight AI tools were integrated into the intervention: Grammarly, ChatGPT, QuillBot, Google Voice Typing, WordQ, Deepseek, DocsBot AI, and Poetry Generator. A mixed-methods design was employed, combining pre- and post-test performance assessments, structured classroom observations, semi-structured educator interviews, and feedback logs. Results demonstrated a 163% average improvement in overall writing skills (pre-test M = 20.4 to post-test M = 52.6), with domain-specific gains of 213% in paragraph writing, 200% in English language proficiency, 158% in functional writing, and 120% in sentence writing. Qualitative findings revealed that AI tools fostered emotional safety, scaffolded independence, multimodal accessibility, and real-time self-regulatory feedback, while implementation challenges included cognitive overload, over-reliance on AI-generated content, infrastructural constraints, and the absence of institutional policy frameworks. This study contributes a theoretically grounded and empirically validated DBR-based instructional model for integrating AI technologies in special education writing instruction, with significant implications for inclusive pedagogy, assistive technology design, and educational policy.
A quantitative approach of its effect on visually impaired junior secondary school students in Ogun state, Nigeria.
Johnson Idowu Oguntola1; Akorede Ayoola Asanre2
Maintaining mathematical knowledge over time can be quite difficult for many learners, particularly for students with visual impairments who often struggle with the abstract nature of mathematics and inability to access the right instructional resources. Hence, this study explores how Seeing AI strategy helps visually impaired junior secondary school students in Ogun State, Nigeria, retain mathematical knowledge. Additionally, it investigates gender impact on students' retention when using this strategy. A quantitative quasi-experimental design was employed, involving a pre-test, delayed post-test, and control group. Two special schools were selected through purposive sampling, with a total of 20 students taking part in the study. Ten of these students were placed in the experimental group, while the other ten made up the control group. Data was gathered using the Mathematics Retention Test (MRT), which had a reliability coefficient of 0.82. The treatment lasted for three weeks. Data collected was analyzed using Analysis of Covariance (ANCOVA). The results indicated that students who engaged with the Seeing AI strategy had significantly better retention in Mathematics compared to those who learned through traditional methods (p<0.05). While gender alone did not have a major impact on retention, there was a significant interaction between the teaching method and gender, highlighting different retention trends among male and female students. The study concludes that Seeing AI improves mathematical retention of visually impaired learners through supportive instruction. Integrating assistive AI tools such as Seeing AI into Mathematics instruction is recommended to create an inclusive and sustainable learning environment.
Isiaka Taiye Owonwami1; Kuburat Mudashir Omowumi2
The emergence of Artificial Intelligence (AI) in education has significantly transformed teaching and learning globally by addressing both academic and psychosocial needs of learners. In North-East Nigeria, persistent insecurity and violence have disrupted schooling and exposed children to traumatic experiences that hinder learning, social interaction and emotional development. Although Social and Emotional Learning (SEL) is widely adopted in Canada, the European Union, the MENA region, other Global South contexts to foster self-awareness, self-management, social awareness and responsible decision-making, its implementation in fragile settings remains limited due to shortages of trained teachers and weak psychosocial support systems. This study therefore evaluates how AI can facilitate SEL in primary schools in North-East Nigeria, with comparative global perspectives. The study involved pupils, teachers and school administrators from selected primary schools in Borno, Adamawa and Yobe States. AI tools were used to classify learners' wellbeing profiles and generate personalized recommendations. AI-enabled SEL activities enhanced self-management, empathy, interpersonal relationship and responsible behaviour, while teacher dashboards provided real-time data for instructional planning and intervention. Findings revealed that AI integration positively influenced learners' emotional wellbeing, motivation, communication, and academic readiness, consistent with evidence from Europe and North America on AI-supported personalized learning and SEL. It also improved teachers' understanding of learners' needs and classroom responsiveness. However, challenges such as data privacy concerns, algorithmic bias, weak infrastructure, and limited accessibility in fragile regions were identified. The study concludes that AI can strengthen SEL when applied ethically, contextually and recommends responsible integration to improve learning outcomes and learner wellbeing in conflict-affected educational systems
P. Selvakumar1; M. Sivaraja2
Artificial Intelligence (AI) is increasingly playing a significant role in environmental research, sustainability initiatives, and ecological protection by providing innovative solutions to complex environmental challenges. Rapid industrialization, climate change, biodiversity loss, pollution, deforestation, and resource depletion have intensified the need for advanced technologies capable of supporting sustainable development and environmental conservation. AI technologies, including machine learning, deep learning, data analytics, remote sensing, and predictive modeling, are transforming the way environmental data is collected, analyzed, and utilized for informed decision-making and ecological management. In environmental research, AI facilitates the analysis of massive datasets generated from satellites, sensors, drones, and climate monitoring systems. These technologies enable scientists to monitor environmental changes in real time, predict natural disasters, and assess ecological risks with greater accuracy and efficiency. AI-driven climate models help researchers understand weather patterns, rising global temperatures, carbon emissions, and extreme climatic events such as floods, droughts, hurricanes, and wildfires. By improving forecasting accuracy, AI supports disaster preparedness and risk mitigation strategies, thereby minimizing environmental and economic damage. AI also contributes significantly to sustainability practices across industries and communities. Smart energy management systems powered by AI optimize electricity consumption, enhance renewable energy integration, and improve energy efficiency in buildings, transportation, and industrial operations. In agriculture, AI-driven precision farming techniques help farmers monitor soil conditions, optimize irrigation, detect crop diseases, and reduce the excessive use of fertilizers and pesticides. These innovations support sustainable food production while conserving natural resources and reducing environmental pollution. Similarly, AI applications in waste management assist in recycling optimization, waste sorting automation, and efficient resource utilization, contributing to circular economy initiatives.
Reimagining artificial intelligence through indigenous knowledge systems among the tribes of Jharkhand
Shyamali Banerjee1
Today's discourse regarding Artificial Intelligence (AI) is still heavily influenced by Eurocentric ideas of technological progress, data collection, and computational reasoning. While several thinkers, including N. Katherine Hayles, Donna Haraway, Kate Crawford, and Yuk Hui, have challenged the relation between culture, humanity, and technology, little has been said about how the indigenous epistemologies could contribute to a shift in the fundamental philosophy of AI. Thus, in this paper, the idea of "Forest Cosmotechnics" is introduced as a new approach to human-technology relations that relies upon the worldviews of the Jharkhand's tribal populations – the Mundas, Oraons, Santhals, and Hos. Relying on the concept of cosmotechnics by Yuk Hui that considers technologies to be rooted within cultural cosmologies and not universal rationalism, this essay discusses how the ecological wisdom of tribes inherent in their oral traditions, Sarna, seasonal festivals, agriculture, and natural resource management provides an entirely new understanding of intelligence. In contrast to machine learning techniques grounded on concepts such as exploitation, optimization, and prediction, tribal knowledge systems operate using principles of reciprocity, relationality, collective memory, and equilibrium with nature. Based on an analysis of stories, folktales, songs, and contemporary writings of tribal communities like poems by Anuj Lugun, the essay attempts to demonstrate that these practices constitute a form of distributed intelligence. Moreover, this research addresses the growing use of digital technology to document Indigenous cultures through archiving, databases, and language technologies. On the one hand, these initiatives have been used to protect endangered Indigenous languages, but at the same time, they pose a risk by turning living cultures into information ready for extraction. Referring to Kate Crawford's critical perspective on AI infrastructures and Safiya Noble's ideas about algorithmic power, the paper identifies the problematic ethical questions regarding digital documentation of Indigenous cultures. The authors call for the creation of culturally sensitive AI systems that treat tribal peoples as epistemic collaborators in technology development. Through the integration of post humanist theory, digital humanities, indigenous studies, and AI ethics, this paper attempts to contribute to ongoing discussions about decolonial technology and argues for the inclusion of the tribal knowledge systems of Jharkhand as an important source for the construction of alternative imaginaries of artificial intelligence. Ultimately, the study posits the need to move away from universal AI and towards culturally specific forms of intelligences.
Examining public services, healthcare and indigenous language preservation
Victor Olusoji1; Nkemdilim Iheanachor2
Although artificial intelligence has the potential to drive positive change and spark innovative product growth and development, such advances carry enormous risks of reinforcing and expanding existing inequalities amidst the increasing adoption of AI systems in various segments of public life in West Africa (Crawford, 2021; Eubanks, 2018). This paper advocates a comprehensive and cross-disciplinary analysis of inclusive development in the framework of AI, to explore three interrelated areas, namely public service delivery, maternal health care and preservation of indigenous languages. These areas, usually studied independently, have common structural weaknesses: governance gaps, lack of data infrastructure, digital exclusion, and a centralisation of knowledge and decision-making with AI in institutions from the Global North. Based on an intercountry comparison (Nigeria, Ghana and Senegal) using a multi-case study design that integrates data from all three countries, this paper aims to analyse how AI is being adopted, controlled and experienced by its users across contexts, with a focus on identifying shared trends and distinguishing similarities and differences in ethical, institutional and distributional implications. There is a steady momentum of tension between technocratic optimism in seeing AI as a development solution and critical concern that it could reproduce some of the structural inequalities, governance failures and epistemic marginalisation that keep occurring (Couldry & Mejias, 2019; Joshi et al., 2020). This paper presents an integrative framework that integrates several issues related to AI's impact on West Africa's development, including equity-oriented design, democratic governance, epistemic justice of West African peoples' knowledge, institutional capacity-building, and regional solidarity based on cooperation among West African countries for shared AI infrastructure, as well as data governance and regulatory harmonisation
Maria Partida1
Artificial intelligence (AI) is rapidly transforming education and the workforce, creating both opportunities and challenges for students as they prepare for academic success, career development, and lifelong learning. As AI technologies become increasingly integrated into educational and professional environments, institutions must ensure that all learners, including neurodivergent students and students with disabilities, have equitable opportunities to develop the knowledge, skills, and competencies needed to succeed in an evolving society. This paper explores how inclusive pedagogy, assessment practices, and institutional change can support workforce readiness among high school dual-enrollment, community college, and university students. Particular attention is given to the ways AI can enhance accessibility, personalize learning experiences, and support the development of communication, critical thinking, and problem-solving skills. At the same time, the paper considers challenges related to equity, digital literacy, ethical AI use, and access to emerging technologies. Drawing on principles of inclusive education and workforce development, the paper proposes a framework for integrating AI literacy and career readiness into educational practice. The framework emphasizes adaptable learning environments, accessibility-focused support systems, and opportunities for students to engage with AI critically, responsibly, and effectively. The paper argues that educational institutions have a responsibility to ensure that AI adoption advances inclusion rather than exclusion. By aligning pedagogy, assessment, and workforce preparation efforts, institutions can better support diverse and neurodivergent learners while preparing them to navigate and contribute to an increasingly AI-driven world.
Integrating AI, circular economy, and smart systems for resource efficiency
Pawanjeet Kaur1; Shailendra Singh2
At present, the digital revolution is bringing changes to our surroundings and providing opportunities to improve sustainability. This is an era of smarter technologies like Artificial Intelligence, Internet of Things, data analytics, blockchain, data-driven smart systems, and green manufacturing, which is extensively used for efficient resource usage, evidence-based decision-making during manufacturing and deployment of processes, reforming agriculture, promoting eco-friendly materials, and supporting circular economy models in sustainable cities. These innovations have demonstrated that science and technology can collaborate to address critical environmental, economic, and social challenges. However, sustainability extends beyond technological progress; it requires alliance with multidisciplinary fields like energy, industries, governance, research, academia, agriculture, and collective responsibility and adaptive policies. There is a need for the future to cultivate critical thinking and responsible innovation for long-term impact through skill-based and experiential education. This paper emphasizes the role of green and smart manufacturing in reducing environmental stress and providing insights on the practical applicability of digitalization in agriculture, energy, education, and smart cities, the challenges observed in the digital era, future tendencies, and how we can all contribute towards a greener and tech-enabled future.
A Design Framework for AI Generated Cultural Twins
Sandra Hodaifa
As artificial intelligence becomes more involved in cultural and educational spaces, there is a growing need for systems that respect the communities whose traditions and knowledge shape our world. This paper introduces the idea of AI cultural twins—synthetic, ethically designed models that mirror the structure of cultural practices without copying or imitating real heritage. Instead of reproducing sacred rituals, private stories, or community‑owned expressions, cultural twins offer a safe, respectful way to explore cultural knowledge through technology. The work focuses on designing AI systems that create these cultural twins using clear ethical boundaries, cultural sensitivity, and human‑centered design. The goal is to build technology that supports learning, museums, and cultural programs without crossing lines of appropriation or misrepresentation. By using abstraction layers, symbolic modeling, and community‑guided parameters, cultural twins become a tool for education and preservation while still protecting cultural identity. This paper argues that cultural twins represent a new direction in AI design—one that values responsibility as much as innovation. They allow us to use technology to keep cultural knowledge alive, but in a way that honors the people behind it. The study outlines practical design principles that developers and educators can use when creating AI systems that interact with cultural content, making cultural twins a promising step toward more respectful and socially aware technology.
Transposing the EU AI Act in the Western Balkans
Andrej Mitić
The EU AI Act is expected to reproduce the "Brussels effect" (Bradford) that turned the GDPR into a global standard. The literature describes this effect as market-driven. This paper argues that in EU candidate countries the mechanism is different: the AI Act is adopted by accession obligation, often before the institutions needed to apply it exist. This could be called a pre-accession Brussels effect, in which the legal text travels faster than regulatory capacity.
The GDPR offers a precedent. Data protection laws in Serbia and North Macedonia, closely modelled on it, show the limits of near-literal transposition and the gap between law in the books and law in action. The AI Act raises the stakes, as it depends on harmonised standards, notified bodies and market surveillance authorities that candidate countries largely lack. It is also a moving target: with the Digital Omnibus (2026), the EU itself postponed most high-risk obligations to 2027 and 2028.
The paper compares Slovenia and Croatia, which apply the Act directly as Member States, with Serbia, which is preparing the region's first comprehensive AI law, and with Montenegro, North Macedonia and Albania, which have no AI-specific legislation yet. The paper argues for a flexible model of reception that distinguishes text transfer from capacity transfer: alignment of the core rules and implementation synchronised with the EU timeline.
Rohan Rajoriyal1; Shuchita Mudgil1; Takaaki Fujita2
In this research, we progress a complete machine learning outline for automatic monitoring and examination of water quality, addressing a significant tests in eco-friendly data science then maintainable organization of water resources. The study utilize numerous datasets composed from IoT-enabled water-quality sensor. The study usages a multi-pronged algorithmic approach with Isolation Forest for irregularity discovery, Random Forest for water-quality organization, K-Means clustering for outline recognition, and Linear Regression for forecast. A main conclusion of the study is that discrepancies in sensor data components, in specific mislabeling of µS/cm and mS/cm, can sternly cooperation the reliability of reasonable significances, which highlight the necessity for severe data authentication then preprocessing prior to machine learning study. The Random Forest classifier attained 92% accurateness in astute amid "Good" and "Bad" water-quality situations. K-Means clustering might classify three dissimilar water-quality outlines with a silhouette score of 0.73. The first effort at irregularity identification was threatening, through together accuracy and recall at 0%, exemplifying the implication of properly parameterizing Isolation Forest, particularly on minor dataset. Regression analysis also indicated the considerable effect of outliers, In addition to its technological contributions, the proposed framework has important implications for environmental and sustainability education, AI-assisted environmental decision-making, and public consciousness. By combining IoT sensing, machine learning and user-friendly data analysis, the framework provides students and educators with real experiences to learn about data-driven water-resource management and prudent use of artificial intelligence for environmental problems. Conversion of sensor data to useful water-quality information can enable researchers, water-resource managers, policy-makers and communities in making more informed environmental decisions. In addition, instant access to water-quality data can increase public awareness of water contamination, conservation, and sustainable resource management. The open-source framework provides a reproducible model for environmental scientists and water-resource administrators, and illustrates the potential role of AI and IoT technologies in furthering scientific progress and encouraging public engagement on sustainability issues. .
Preetha Kamath B1; Saleem S Tevaramani1; Priyadharshini V1; Ramya K R1; Pragati Pukkela1
Electric vehicles (EVs) are becoming a primary solution for sustainable transportation due to reduced greenhouse gas emissions and improvement in energy efficiency. The performance interaction of subsystems, including motor controllers, advanced battery management systems (BMS), and robust charging infrastructures. The paper proposes EV technologies, ranging from machine learning-based battery diagnostics and nonlinear reliability models to wireless power transfer (WPT) systems and solar-assisted drives. This research establishes critical thermal benchmarks for BMS by demonstrating that Li-ion battery voltage levels degrade aggressively at a 2C-rate compared to a 0.3C-rate, when ambient temperatures reach 40°C. It validates the use of Explainable AI (XAI) through LightGBM and XGBoost as a powerful tool for grid management, a substantial 63% to 89% reduction in peak regional energy demand by optimizing charging schedules across different locations and times, without the need for physical infrastructure. In terms of motion control, integrated convex optimization is highly effective for eco-driving. By treating road geometry, including slope and curvature as active variables, our approach reduced the computational load to less than 0.005%. This breakthrough allows the energy saving algorithms to run in real-time. The evolution between 2023-2025 transitions toward 300 kW ultra-fast charging systems complements the standard battery capacities of 75–80 kWh. GREET model for a comprehensive life-cycle analysis, optimizes vehicle-to-grid (V2G) strategies drives a 28% decrease in total carbon emissions by 2030. This provides a strategic blueprint for balancing performance, longevity, and ecological impact in the next generation and analysis results will be shared to the user through Email and messages.
Review of privacy-preserving approaches for behavioral biometrics to ensure data confidentiality
Preetha Kamath B1; Electa Alice Jayarani Appadurai1
Federated Learning has attracted significant interest in recent years for its ability to keep data localized, thereby enhancing privacy and security while still enabling the development of robust, generalized models. Ensuring the confidentiality of sensitive health information, such as ECG recordings, is critically important. In this context, Federated Learning serves as a promising approach to safeguard privacy during data classification. This study explores this potential by comparing centralized learning models with Federated Learning models trained on both Independently and Identically Distributed (IID) data and Non-IID data. The result of this study would open path for conduct further experiments on various observable behaviors, where pedometric data obtained through federated learning can play a more significant role in gait analysis for financial fraud detection.
Integrating knowledge distillation, self-play, recursive generation, and AutoML for model-to-model instruction
Joseph Aref1
AI systems now process data at scales and speeds that surpass human comprehension, yet the data they consume often suffers from four critical deficiencies: corruption, redundancy, bias, and insufficiency. These problems compound as models are deployed in real-world settings, leading to wasted computational resources, unreliable outputs, and systemic failures. This paper proposes a novel Automated Training Loop that integrates four existing AI training paradigms—Knowledge Distillation, Self-Play, Recursive Training, and Autonomous Machine Learning (AutoML)—into a coordinated three-stage feedback loop. In this framework, a Teacher model generates training data with high-level guidance; a Student model filters, validates, and requests targeted data augmentation; and a Final Product model is trained by the Student with minimal human intervention. We demonstrate that this approach reduces data storage requirements by 35%, improves classification accuracy by 11.3 percentage points compared to static synthetic data, and achieves 91.2% of real-data baseline performance. The framework successfully trained a team of AI models to complete a set of object recognition tasks with high precision. These findings suggest that fully automated, model-to-model instruction is feasible and offers significant efficiency gains for AI development, with direct applications in mechatronics and autonomous systems.
Abdul-Aziz Al-Izzo1; Abduljalil Najjar2
Traditional pharmacology education has historically relied on live-animal experimentation to demonstrate the physiological effects of drugs on organ systems. This conventional approach carries significant logistical burdens, including high procurement and husbandry costs, rigid scheduling constraints that limit repeated hands-on practice, and unpredictable biological variability that can compromise learning objectives. More importantly, ethical concerns surrounding animal welfare have become increasingly prominent in medical education, prompting a global shift toward the 3Rs framework — Replacement, Reduction, and Refinement — first articulated by Russell and Burch in 1959. This paper presents a comprehensive literature review evaluating the viability of Virtual Reality (VR) and computer-based simulation as an ethical and pedagogically effective alternative to in-vivo pharmacology training. Drawing on peer-reviewed evidence from medical, nursing, and pharmacology education research published between 2017 and 2026, the review synthesizes findings on student perceptions, measurable learning outcomes, and operational effectiveness. The evidence indicates that VR-based pharmacology platforms produce moderate-to-large improvements in knowledge acquisition and practical skill development relative to traditional laboratory methods, while simultaneously reducing students' reported discomfort and fear associated with handling live subjects. From a logistical standpoint, the recurring cost of maintaining simulation software is substantially lower than the cost of breeding, housing, and disposing of experimental animals. However, the transition to immersive VR introduces its own ethical and physical challenges. Head-mounted display use is associated with cybersickness — a cluster of symptoms including nausea, disorientation, and oculomotor strain — and immersive systems routinely collect granular biometric data such as eye-gaze, motor response, and physiological signals, raising unresolved questions about informed consent, data ownership, and psychographic profiling. Accessibility remains uneven, with infrastructure costs and digital-divide effects limiting adoption in low- and middle-income countries. This paper concludes that VR offers a flexible, scalable, and ethically preferable alternative that modernizes pharmacology training, but that its responsible implementation requires deliberate attention to cybersickness mitigation, privacy-by-design data governance, and equitable access, particularly in resource-constrained educational settings such as those found across the Levant region.
Ljubinko Stojanović1; Petar Spalević1
This paper examines the application of error control coding techniques in Li-Fi systems with the aim of improving the reliability and efficiency of optical wireless communications. Starting from the specifics of the optical channel with intensity modulation and direct detection (IM/DD), the main sources of signal degradation are analyzed, including ambient noise, thermal noise, photodetector noise, and limitations of optical transmission in indoor environments. The focus is placed on the application and comparative analysis of different coding techniques, including block codes, Reed–Solomon codes, BCH codes, convolutional codes, and LDPC codes, in terms of their ability to reduce the probability of errors in data transmission. The paper considers the impact of these techniques on key performance metrics such as bit error rate (BER), coding gain, latency, decoding complexity, and system energy efficiency. It also analyzes the compatibility of the selected coding techniques with Li-Fi system requirements, especially in scenarios with high data rates, limited hardware resources, and increased sensitivity to noise and interference.
Ljubinko Stojanović1
This paper considers the problem of inter-cell interference in dense Li-Fi attocell networks and proposes an approach based on adaptive cell formation and an interference-aware design. Starting from the specifics of the optical channel with intensity modulation and direct detection (IM/DD), different cell structures are analyzed, including regular, frequency-separated, merged multi-access-point cells, virtual transmission, and user-oriented cells. Special focus is placed on modeling line-of-sight (LOS) interference, overlap of attocell zones, and the impact of LED deployment density on communication quality. The paper introduces an adaptive framework that enables dynamic adjustment of cell boundaries and resource allocation depending on user position and interference levels. The performance of the proposed approach is evaluated through metrics such as signal-to-interference-plus-noise ratio (SINR), throughput, and spectral efficiency. Adaptive cell formation can significantly reduce interference at cell edges and improve overall system performance compared to static approaches.
Molecular mechanisms and future prospects
Madhu Kushwaha1; Pankaj Tiwari1
Cancer remains one of the foremost causes of morbidity and mortality worldwide, posing a substantial challenge to healthcare systems despite remarkable advances in diagnosis and treatment. Although chemotherapy, radiotherapy, immunotherapy, and targeted therapies have improved patient outcomes, their clinical application is frequently constrained by adverse effects, multidrug resistance, limited selectivity, and high treatment costs. These limitations have intensified the search for safer, multi-targeted, and biologically effective therapeutic agents from natural sources. Medicinal plants continue to play a pivotal role in drug discovery because they contain structurally diverse bioactive molecules with broad pharmacological activities and favorable safety profiles. Among these, the Solanaceae family has emerged as an important source of phytochemicals exhibiting significant anticancer potential. Numerous secondary metabolites isolated from Solanaceae species have demonstrated the ability to inhibit cancer progression by regulating apoptosis, cell proliferation, angiogenesis, metastasis, oxidative stress, inflammation, and multiple intracellular signaling pathways. This review critically examines the current evidence on the anticancer potential of Solanaceae-derived bioactive compounds while emphasizing the growing contribution of Artificial Intelligence (AI) to natural-product-based drug discovery. A comprehensive analysis of studies published between 2011 and 2025 was performed using major scientific databases, including PubMed, Scopus, Web of Science, Google Scholar, Wiley Online Library, and Semantic Scholar. The collected literature was systematically evaluated to summarize phytochemical composition, pharmacological activities, molecular targets, mechanisms of action, and therapeutic prospects of Solanaceae species in cancer management. Beyond conventional experimental findings, this review highlights how AI-driven technologies are reshaping phytopharmaceutical research. Advanced computational approaches, including machine learning, deep learning, quantitative structure–activity relationship (QSAR) modeling, molecular docking, molecular dynamics simulations, network pharmacology, predictive toxicology, and multi-omics data integration, are increasingly being employed to identify promising phytochemicals, predict biological targets, optimize lead compounds, evaluate pharmacokinetic characteristics, and estimate safety profiles. These AI-enabled methodologies substantially reduce the time, cost, and complexity associated with traditional drug discovery while improving the efficiency of identifying therapeutically relevant natural compounds. The review compiles evidence on approximately eighteen Solanaceae species reported to possess anticancer activity, encompassing nearly sixty-three bioactive phytochemicals with demonstrated efficacy in preclinical investigations. Their mechanisms involve modulation of key oncogenic pathways, including PI3K/Akt/mTOR, MAPK, NF-κB, JAK/STAT, Wnt/β-catenin, and p53 signaling, resulting in suppression of tumor growth, induction of programmed cell death, inhibition of metastatic progression, and regulation of the tumor microenvironment. Comprehensive tables summarizing plant sources, major phytochemicals, molecular targets, experimental evidence, and AI-assisted computational predictions are included to facilitate comparison and future investigation. Furthermore, this review identifies current knowledge gaps and discusses future opportunities for integrating AI with phytochemistry, systems biology, precision oncology, and translational pharmacology. The convergence of artificial intelligence with natural-product research provides an innovative framework for accelerating the discovery and development of next-generation anticancer agents derived from the Solanaceae family. By combining computational intelligence with experimental validation, AI has the potential to transform traditional medicinal knowledge into evidence-based precision therapeutics, ultimately contributing to safer, more effective, and personalized cancer treatment strategies.
Ammar Yaser Fashtouk1
LLMs (Large Language Models) excel at general-purpose benchmarks like MMLU and GSM8K. However, these benchmarks do not test domain-specific mechatronics reasoning including kinematics, PID control, sensor fusion, and state-space modeling. No systematic evaluation of LLMs on authentic mechatronics engineering problems currently exists. This study evaluates four LLMs (Claude 3.5 Sonnet, GPT-4, Gemini 1.5 Pro, and Llama 3 70B) on 130+ undergraduate mechatronics exam questions collected from publicly available curricula. Questions span six categories: kinematics, PID tuning, sensor fusion, circuit analysis, state-space control, and dynamics. Each model receives identical zero-shot prompts. Responses are graded blindly against a detailed rubric and compared to human student performance baselines. Pilot results indicate LLMs achieve only 55–68% accuracy versus 78–85% for humans. Critical failure modes include sign errors in rotation matrices, misunderstanding of integral windup in PID controllers, and poor explianing about sensor noise covariance. This work establishes the first mechatronics reasoning benchmark for LLMs, revealing significant gaps that general benchmarks miss. These findings directly inform AI adoption in engineering education and industry.
Yogesh Virupaksha Chimate1; Varsha Prashant Desai2
Food adulteration poses significant health risks and economic losses worldwide, particularly in spices such as chilli powder. This study proposes an artificial intelligence-based system for automated detection and quantification of adulteration levels in chilli powder using image processing and machine learning techniques. A structured dataset comprising 480 images across 12 adulteration levels (0% to 100%) was developed. Feature extraction was performed using color histograms, Local Binary Patterns (LBP), and Gray-Level Co-occurrence Matrix (GLCM), resulting in a 526-dimensional feature vector. Three learning paradigms were explored: binary classification, multiclass classification, and regression-based estimation. The results demonstrate that while binary classification achieved near-perfect accuracy (up to 100%), multiclass classification performance decreased (~78%), highlighting inter-class similarity challenges. Regression-based estimation using XGBoost achieved a Mean Absolute Error (MAE) of 4.42%, demonstrating superior capability in predicting continuous adulteration levels. Analytical evaluation using confusion matrices, error distributions, PCA visualization, and tolerance accuracy curves reveals that regression provides a more robust and practical solution for real-world applications. The proposed system offers a scalable and cost-effective approach for food quality monitoring.
Anshul Jain1; Madhu Kushwaha2
Assessing and forecasting the condition of freshwater bodies pose significant difficulties, especially amid escalating climate disruptions projected ahead. These obstacles stem in part from intricate mechanisms that influence water conditions, combined with labor-intensive and costly monitoring efforts that worsen data shortages. Conventional mechanistic and empirical models frequently underperform in estimating water states. This overview suggests that neural network-based advanced learning offers a valuable yet underexplored method for decoding complex patterns within voluminous datasets. We illustrate how these advanced techniques can mitigate data limitations by bridging gaps in time and location, while also supporting the creation and validation of theories through pinpointing key factors affecting water conditions. The discussion weighs the advantages and drawbacks of these innovative strategies against established methods, emphasizing their promise as a vital tool for tackling obstacles and uncovering fresh insights in the field of water condition research.
An overview of structures and implementations
Anshul Jain1; Madhu Kushwaha2
City Virtual Models (CVMs), also known as digital twins or virtual replicas of urban environments, are rapidly emerging as transformative tools for enhancing the resilience and sustainability of essential systems (ES)—such as energy grids, transportation networks, water supply, waste management, and critical infrastructure—in smart cities. This is particularly crucial amid escalating threats from environmental shifts, including climate change-induced extreme weather events, rising temperatures, flooding, and resource scarcity. By integrating real-time data streams from IoT sensors, advanced simulation techniques, predictive analytics, AI-driven forecasting, and big data processing, CVMs enable data-informed decision-making, optimize urban operations, simulate "what-if" scenarios, and facilitate proactive risk management. They allow city planners and administrators to model complex interactions within metropolitan systems, improving efficiency, reducing vulnerabilities, and supporting adaptive responses to dynamic environmental pressures. Given the inherent complexity of urban ecosystems—marked by interconnected networks, diverse stakeholders, and heterogeneous data sources—effective CVM deployment demands careful attention to interoperability, data compatibility, standardization, scalability, and multi-domain integration. This article examines the pivotal role of CVMs in managing ES under climate uncertainty, identifies key threats, highlights cutting-edge technological advancements, and reviews real-world implementations and case studies. Despite their promise, CVM adoption faces significant barriers, including high implementation costs, data privacy concerns, technical silos, cybersecurity risks, and equitable access issues. Ultimately, as CVMs continue to evolve, sustained innovation, ethical deployment, and equitable resource distribution will be essential to maximize their transformative impact, building more resilient, sustainable, and adaptive urban systems capable of withstanding ongoing and future environmental shifts.
A comprehensive analysis
Anshul Jain1; Madhu Kushwaha2
The construction and demolition (C&D) industry generates significant waste, contributing to environmental degradation and resource depletion. Effective waste segregation is critical for sustainable waste management, yet traditional methods are labor-intensive, error-prone, and inefficient. Artificial Intelligence (AI) offers transformative potential in automating and optimizing C&D waste segregation. This paper investigates the role of AI in enhancing the segregation of C&D waste, focusing on machine learning, computer vision, and robotic systems. Through a systematic review of existing technologies, case studies, and experimental data, the study evaluates AI's efficacy in improving classification accuracy, reducing landfill waste, and promoting circular economy principles. The findings highlight AI's ability to achieve up to 95% accuracy in waste sorting, reduce operational costs by 15–20%, and enhance material recovery rates. Challenges such as limited datasets and technological integration are addressed, with recommendations for future research to advance AI-driven waste management.
Devika G1; Asha Gowda Karegowda2
The integrity of official documents such as academic certificates, identity cards, and government-issued letters is in-creasingly threatened by the availability of sophisticated digital editing tools. Existing automated detection methods are either limited to single forensic signals or rely on computationally expensive architectures unsuitable for practical deployment. This paper presents ForensicAI, a multi-modal document forgery detection system that integrates a fifteen-check ensemble forensic engine with a knowledge-distilled EfficientNet-B0 deep learning classifier. The system estimates the authenticity of a submit-ted document by independently computing scores from Error Level Analysis, DCT-based copy-move detection, metadata finger-printing, OCR-driven font consistency analysis, stamp integrity verification, signature ink entropy, and nine additional forensic checks, then aggregating them into a unified forgery score. A key contribution is a knowledge distillation framework that trans-fers forgery-discriminative representations from a Keras-based teacher model trained on handwritten signature verification to a PyTorch student network trained on document images, enabling effective learning from a limited synthetic dataset of one thousand samples. The system is deployed through a lightweight Flask web interface that produces line-by-line forensic explanations alongside ELA maps, noise maps, and annotated visualisations in under three seconds per document. Experimental evaluation on a synthetic benchmark of five document categories yielded a validation accuracy of 100%, an AUC-ROC of 1.000, and an F1-score of 1.000, with the ensemble engine achieving a precision of 0.98 and recall of 0.96 across all check modalities. These results confirm that combining independent forensic signals through ensemble aggregation substantially outperforms any single-modality baseline and produces verdicts that are both accurate and interpretable for non-technical stakeholders.
Devika G1; Asha Gowda Karegowda2
This study aims to enhance disaster management and earthquake risk severity prediction using advanced modelling tools, taking into consideration our present understanding of earthquakes as complex natural phenomena. Predictions of earthquakes, which can range from seconds to decades, are essential for making plans to respond to and prepare for disasters. Both linear and nonlinear forecasting models, specifically the ARIMA and LSTM models, are utilised in this research, which makes use of seismic data collected. The statistical efficiency of ARIMA and the deep learning capabilities of LSTM are combined in a HYBRID model to improve forecast accuracy. The results show that the HYBRID model achieves a success rate of 98.06% in estimating the severity of earthquake risk, which is far better than individual models. This impressive level of accuracy highlights the promise of data-driven, integrated methods for accurate earthquake prediction. In sum, the research adds to what is already known by providing a solid foundation for earthquake prediction that can direct choices for catastrophe management, infrastructure design, and emergency preparedness. To lessen the catastrophic effects of earthquakes on people and economies around the globe, the suggested model highlights the need of hybrid analytics.
Raghad Mahmoud Fallaha1
Medication errors, poor adherence, and incomplete patient history remain major challenges in community pharmacies, affecting treatment outcomes and patient safety. This project proposes an AI-based Smart Pharmacy System that integrates artificial intelligence with a mobile health application to support pharmacists and improve patient-centered care. When patients enter the pharmacy, they interact with an AI-supported interface that collects essential information, including age, weight, medical history, allergies, chronic diseases, and current medications. The pharmacist then enters the dispensed medication, and all data are securely stored. The system generates a unique patient code linked to the mobile application, allowing access to the full medication profile, reminds patients of scheduled dose times, and provides clear instructions through text and AI voice guidance. A key feature of the system is continuity of care. When patients return for new treatment, pharmacists can review previous medications, chronic conditions, past complications, and treatment responses through the patient code. This supports safer and more precise therapy, especially for elderly patients and those with chronic diseases. Another component is medication adherence tracking. The application divides the treatment plan according to the prescribed duration, reminds patients of scheduled dose times, and allows them to confirm each dose. This creates a record that helps pharmacists evaluate whether poor treatment response is due to missed or irregular doses. The system also enables direct communication between patients and pharmacists regarding side effects, questions, or medication-related problems. AI continuously monitors potential drug interactions and dose conflicts when new medications are entered and sends timely alerts to the pharmacist. Overall, the system improves adherence, reduces errors, enhances patient understanding, and supports responsible AI use without replacing pharmacists.
Ncube Mthandazo Josephat Junior1; Sibanda Khulekani2
Preeclampsia remains a significant cause of maternal complications and death, therefore early and accurate prediction is critical to reducing complications. Traditional machine learning models rely on structured data such as age, BMI, and blood pressure, while large language models (LLMs) offer the advantage of processing both structured and unstructured data, such as clinical notes. This study developed and compared the performance of traditional machine learning models (Random Forest, XGBoost) and an LLM (BERT) in predicting preeclampsia, using maternal health data from Mpilo Central Hospital and Kaggle. This study follows the CRISP-DM framework for data preprocessing, model training, and evaluation and utilizes a machine learning-based quantitative research methodology. Data preprocessing using Pandas and NumPy and feature selection using SelectKBest preceded model training and evaluation. Random Forest and XGBoost were implemented with scikit-learn, while BERT used Hugging Face Transformers. Evaluation metrics included accuracy, precision, recall, and F1-score. Result demonstrated BERT as the top performer with an accuracy of 92%, outperforming XGBoost (90%) and Random Forest (88%). LLMs improved predictive performance by leveraging unstructured data alongside structured features. Results show that predictive performance on clinical risk prediction tasks can be improved using LLMs by leveraging unstructured data, and they are promising for healthcare decision-making improvement.
P. Selvakumar1; M. Sivaraja2
Artificial Intelligence (AI) has emerged as a transformative force in healthcare and medical research, revolutionizing the way diseases are diagnosed, treated, monitored, and prevented. By integrating advanced computational algorithms, machine learning, deep learning, natural language processing, and predictive analytics into healthcare systems, AI is enabling faster decision-making, improved patient outcomes, and enhanced operational efficiency. The increasing availability of healthcare data from electronic health records, wearable devices, genomic sequencing, and medical imaging has further accelerated the adoption of AI-driven solutions across clinical and research domains. In healthcare delivery, AI technologies are significantly improving diagnostic accuracy and precision medicine. AI-powered imaging systems can detect abnormalities in radiology, pathology, and dermatology with remarkable accuracy, often matching or surpassing human experts in specific tasks. Predictive analytics models assist clinicians in identifying disease risks, forecasting patient deterioration, and recommending personalized treatment plans based on patient-specific characteristics. Virtual health assistants and AI chatbots are also enhancing patient engagement by providing continuous monitoring, medication reminders, and real-time health guidance. Furthermore, robotic-assisted surgeries supported by AI contribute to minimally invasive procedures, reducing recovery time and improving surgical outcomes. AI is also transforming medical research by accelerating drug discovery, clinical trials, and biomedical innovation. Traditional drug development processes are often time-consuming and costly; however, AI algorithms can rapidly analyze molecular structures, predict drug interactions, and identify potential therapeutic compounds.
Properties, estimations and applications
Umar Muhammad Adamu1; Suleman, I2; Zakariyau, N. R3
In this paper, a novel univariate continuous lifetime distribution called the Novel Modified New Exponential–Gamma Distribution (NMnEGD) is proposed as a flexible extension of the New Exponential–Gamma family. The proposed distribution is constructed by modifying the New Exponential–Gamma distribution using a generalized gamma–type mechanism, thereby introducing additional shape parameters that enhance modeling flexibility. Several well-known lifetime distributions, including the Exponential, Lindley, Weibull, Gamma, Power Lindley, and Generalized New Exponential–Gamma distributions, are obtained as special cases of the proposed model. The mathematical properties of the NMnEGD are thoroughly investigated, including the probability density function, cumulative distribution function, moments, moment generating function, order statistics, and important reliability characteristics such as the survival function, hazard rate, mean residual life, and mean waiting time functions. Parameter estimation is carried out using the maximum likelihood, least squares, weighted least squares, and Cramér–von Mises estimation methods. The performance of the proposed model is illustrated using a real-life dataset on the strengths of glass fibers. Goodness-of-fit comparisons based on the Akaike Information Criterion, corrected Akaike Information Criterion, Bayesian Information Criterion, and −2 log-likelihood demonstrate the superior fitting ability of the NMnEGD over several existing competing distributions. These results highlight the usefulness of the proposed distribution for modeling lifetime data with diverse hazard rate behaviors.
A bibliometric analysis (1999–2026)
Basavaraju1; Nagaraja Y2
Volleyball is a globally popular sport characterized by repetitive jumping, landing, spiking, and blocking actions that increase the risk of musculoskeletal injuries, particularly anterior cruciate ligament (ACL) injuries and other lower-limb disorders. The rapid growth of scientific literature in sports injury epidemiology and biomechanics has created challenges in identifying emerging trends and knowledge structures. This study demonstrates the application of artificial intelligence-assisted bibliometric techniques to map the global research landscape related to volleyball injury epidemiology and biomechanical risk factors from 1999 to 2026. Bibliographic data were retrieved from the Scopus database using keywords related to volleyball, injury, biomechanics, kinematics, and kinetics. Data analysis was performed using the Bibliometrix package and Biblioshiny platform, which employ AI-supported visualization and knowledge-mapping approaches to identify influential publications, authors, journals, thematic clusters, and emerging research directions. The findings reveal a substantial increase in publication output after 2014, reflecting growing scientific interest in injury prevention and biomechanical assessment. Keyword network analysis identified major themes related to knee injuries, ACL risk factors, landing mechanics, and sports injury prevention. Thematic mapping indicated a shift toward evidence-based injury prevention strategies, longitudinal studies, and technology-enhanced biomechanical assessment. Furthermore, AI-assisted bibliometric visualization enabled efficient exploration of complex research networks and facilitated the identification of knowledge gaps and future research priorities. The study highlights the growing role of artificial intelligence in scientific knowledge organization and research evaluation. It demonstrates how AI-supported analytical tools can assist researchers, educators, and policymakers in understanding evolving trends in sports science and promoting data-driven decision-making in education and research.
Mohamad Rami Al Jundi1
Artificial Intelligence (AI) is increasingly transforming manufacturing systems through predictive analytics, process optimization, and intelligent decision-making. In materials engineering and metallurgy, machine learning models have demonstrated significant potential for predicting material properties such as hardness, tensile strength, wear resistance, and impact toughness based on processing parameters and microstructural characteristics. However, the adoption of AI-driven solutions in industrial and educational environments remains limited due to the "black-box" nature of many machine learning algorithms, which often provide predictions without transparent explanations. This lack of interpretability creates challenges for engineers, educators, and students who require a clear understanding of the relationships between manufacturing variables and material performance. This paper explores the role of Explainable Artificial Intelligence (XAI) as a framework for enhancing transparency, trust, and educational value in smart manufacturing systems. A conceptual model is proposed for integrating XAI techniques into materials engineering and mechatronics education, enabling learners to visualize and understand how process parameters influence predicted material properties. The framework combines machine learning prediction models with explain ability methods to support both industrial decision-making and engineering learning outcomes. Particular attention is given to applications involving heat treatment processes and advanced metallic materials, where accurate prediction and interpretation of property changes are critical. The study argues that explain ability should be considered not only a technical requirement but also an ethical and educational necessity in the era of Industry 5.0. By improving transparency and human understanding, XAI can strengthen collaboration between artificial intelligence systems and engineering professionals while preparing future engineers to responsibly utilize AI technologies
Federated knowledge distillation for hybrid adaptive binarized neural networks
K T Dhanasekaran1; E.Anbalagan2
—the rapid growth of edge intelligence and Internet of Things (IoT) systems has intensified the need for deep learning models that are both lightweight and communication-efficient, especially in environments constrained by limited resources and strict privacy requirements. While Federated Learning (FL) enables decentralized training without raw data sharing, traditional FL approaches often suffer from high communication costs and degraded performance under heterogeneous, non-IID data distributions. Similarly, Binarized Neural Networks (BNNs) significantly reduce computational and memory requirements through binary weights and activations, yet they remain prone to instability and limited representational capacity in distributed optimization.. To overcome these challenges, we introduce Federated Knowledge Distillation–enhanced Hybrid Adaptive Binarized Neural Networks (FKD-HABNet), a unified framework that integrates hardware-aware binarization with communication-efficient federated optimization. In this approach, compact binarized student models collaboratively learn from globally aggregated teacher logits via Federated Knowledge Distillation, thereby reducing bandwidth requirements while enhancing generalization across diverse clients. The framework further incorporates Adaptive Long-Tailed Activation Binarization (ALAB), Pre-activation Guided Binarization (PGB), Reduced Approximate Stochastic Depth (RASD), probabilistic gradient pruning, and adversarial dual batch normalization to improve convergence stability, robustness, and feature representation. Extensive experiments on CIFAR-10 and ImageNet benchmarks using ResNet and VGG architectures demonstrate that FKD-HABNet consistently achieves superior accuracy, communication efficiency, and inference speed compared to conventional BNN and federated approaches. By significantly lowering communication overhead and memory consumption while maintaining strong performance under non-IID conditions, FKD-HABNet provides a scalable and practical solution for next-generation edge intelligence applications.
Zainab Jomaa Hadid1
Toward Faster, Smarter, and More Accurate Healthcare Early detection is one of the most powerful ways to improve patient outcomes, because many diseases are more treatable when they are discovered before reaching advanced stages. Artificial Intelligence offers a promising opportunity to strengthen this process by helping healthcare professionals identify early warning signs quickly, accurately, and efficiently. This presentation focuses on the benefits of using AI in early disease detection. AI systems can analyze large amounts of medical data, including medical images, blood test results, ECG readings, symptoms, and patient history. Through pattern recognition, AI can detect small or hidden changes that may be difficult to notice during the early stages of disease. This can be especially useful in detecting conditions such as cancer, heart disease, diabetes, lung disorders, and neurological diseases. One of the main benefits of AI is speed. By processing medical information in a short time, AI can help reduce delays in diagnosis and allow doctors to make faster decisions. Another important benefit is accuracy, as AI can support doctors by highlighting abnormal patterns, suggesting further testing, and identifying high-risk cases. This can reduce human error and improve the chance of starting treatment earlier. AI can also make healthcare more accessible, especially in hospitals with limited resources or areas where specialists are not always available. It can act as a smart assistant that supports medical teams, rather than replacing them. In conclusion, Artificial Intelligence can play a major role in the future of healthcare by supporting early diagnosis, improving accuracy, saving time, and giving patients a better chance of successful treatment.
Dragan Kreculj1; Đorđe Dihovični2; Nada Ratković Kovačević2; Aleksandra Pavlović2
Digital twin is a dynamic, virtual replica of a physical object, process, or systems. AI (Artificial Intelligence) represents the capability of computer systems to perform tasks that typically require human cognition, as reasoning, problem-solving and learning. It operates by processing large amounts of data through complex algorithms to identify patterns, make predictions and automate processes. AI combined with digital twins creates specific added value. The use of generative AI increases the power of the digital twin technology. By leveraging AI algorithms like machine learning (ML), neural networks and generative AI, these digital, virtual replicas analyze real-time IoT data, predict future states and behaviors, and autonomously optimize physical assets, simultaneously saving time and resources. The listed transforms digital twins from passive 3D model into active, so called autonomous problem-solvers. AI-powered digital twins technology is revolutionizing cloud computing by providing dynamic, intelligent models, that replace real physical systems. By connecting real-time IoT sensor data with ML, AI enables these digital replicas to predict equipment failures, optimize workflows, simulate "what-if" scenarios, dramatically reducing downtime and costs across industries. With ML, as the field od AI, and generative AI systems can continuously learn, predict equipment failures, compare various scenarios and suggest demand optimizations. Agentic AI can independently monitor conditions and execute control signals. Key industry applications of mentioned technology include: intelligent manufacturing, smart cities, robotics, logistics, energy and power grids, aerospace, healthcare.
Aminu Adamu Ahmed1; Ali Usman Abdullahi2; Abdulsalam Yau Gital1; Abubakar Yusuf Dutse1
Deep learning models applied to e-commerce demand prediction routinely achieve excellent training-set performance yet generalize poorly to unseen data - a failure mode known as overfitting that undermines their operational reliability. This study develops and evaluates a hybridized stacking framework that combines Bidirectional Encoder Representations from Transformers (BERT) for context-aware sentiment extraction with a Bidirectional Gated Recurrent Unit (Bi-GRU) for sequential demand forecasting, regularized through dropout, L2 weight decay, mini-batch training, and 5-fold cross-validation. The framework was trained and evaluated on two datasets: a primary survey-derived dataset (n = 114) and a large-scale secondary Kaggle e-commerce dataset (n = 500,000; test n = 75,000). Across both datasets, the hybridized Bi-GRU consistently outperformed standalone GRU and standalone BERT configurations. On the Kaggle dataset, the Bi-GRU achieved 98.86% sentiment-classification accuracy, a Mean Absolute Percentage Error (MAPE) of 2.78% for demand forecasting, and a train-validation loss gap of only 0.0049 - well below the 0.02 overfitting threshold applied in this study - compared with a gap of 0.2012 for standalone BERT on primary data. A systematic dropout-rate sweep identified 0.3 as the empirically optimal regularization rate on both datasets. Benchmarked against two recent comparable studies, the proposed framework reduced MAPE by 16.32 and 24.42 percentage points respectively. Feature-importance analysis further showed that behavioral constructs collectively accounted for approximately 68% of the Bi-GRU's predictive weight, evidencing a substantive theory-to-computation convergence. The findings offer both a validated architecture for overfitting-resilient demand prediction and a replicable regularization protocol for hybrid transformer-recurrent models operating on data-constrained e-commerce environments.
A Review of Lightweight Cryptographic Overhead in Microcontroller-Based UAV Control Loops
Ghazi Zuaieb
Unmanned Aerial Vehicles (UAVs) now depend on cryptography to protect their telemetry and command links from interception and spoofing. But the microcontrollers inside UAVs have very limited resources, and this creates a problem that affects both the software and the physical aircraft. Standard encryption algorithms need a lot of processing power, memory, and energy. This raises power consumption, which calls for heavier batteries, and heavier batteries shorten flight endurance. There is a bigger concern too. When the CPU spends too much time on cryptography, it processes sensor data and motor commands late. Proportional-Integral-Derivative (PID) control loops have to run at high frequencies to keep the aircraft stable, so this delay shows up in the real world as actuator jitter, unstable flight paths, and a higher risk of collisions. This paper reviews Lightweight Cryptography (LWC) algorithms in a systematic way and asks whether they are practical for UAV embedded systems. We bring together recent experimental results and compare the trade-off between security strength and mechatronic performance. The metrics we look at are CPU utilization, memory footprint, bus load, and power consumption in microjoules. The review shows the exact latency levels at which data security starts to harm aerodynamic stability. It also offers a framework for choosing the best LWC solution for the next generation of autonomous flight systems.
Radwa Barakat1; Rawan Zainiyeh2
Plant-mediated synthesis of metallic nanoparticles has gained increasing attention as a sustainable, cost-effective, and environmentally friendly alternative to conventional chemical and physical synthesis methods. This study explores the green biosynthesis of silver nanoparticles (AgNPs) using an aqueous extract of Artemisia leaves as a natural source of reducing and stabilizing phytochemicals. The plant extract was prepared through controlled thermal extraction at 60 °C and subsequently combined with a 1 mM silver nitrate (AgNO₃) solution at a volumetric ratio of 9:1. The synthesis process was conducted under optimized alkaline conditions (pH 9), with continuous magnetic stirring at a controlled temperature of 30 °C. The formation of AgNPs was initially evidenced by a pronounced color transition from light yellow to dark brown or reddish-brown, indicating the development of surface plasmon resonance (SPR) associated with silver nanoparticle formation. Subsequent characterization using ultraviolet–visible (UV–Vis) spectroscopy provided further evidence of successful nanoparticle biosynthesis through the appearance of a characteristic SPR absorption band within the visible region. These observations support the role of Artemisia leaf-derived phytochemicals in facilitating the reduction of silver ions and potentially contributing to nanoparticle stabilization. The proposed biosynthetic approach offers a straightforward, economically viable, and environmentally responsible strategy for producing silver nanoparticles while minimizing reliance on hazardous chemical reagents. The findings highlight the potential of Artemisia-mediated green synthesis as a promising platform for the development of functional nanomaterials, particularly for future antimicrobial and biomedical applications, although further physicochemical characterization and biological activity assessments are required to establish nanoparticle stability, functionality, and suitability for such applications.
How Artificial Intelligence Is Revolutionizing Pharmaceutical Quality Control and Defect Detection
Nour Al Yousef1
Pharmaceutical quality control is essential for ensuring the safety, quality, and reliability of medicinal products. Traditional inspection methods may face challenges related to inspection speed, consistency, and the identification of subtle defects. Artificial intelligence (AI), particularly machine learning and computer vision, offers promising approaches to improving pharmaceutical quality control by analyzing product images and identifying visible defects. This paper explores the role of AI in pharmaceutical quality control, focusing on the detection of tablet defects such as chipping, breaking, color non-uniformity, and surface irregularities. It introduces the basic principles of AI-based inspection systems and discusses how these technologies can support faster and more consistent defect identification. Relevant scientific studies are examined to demonstrate current applications and reported performance. In addition, the presentation addresses important limitations, including data quality, model validation, and the risk of incorrect classifications. Although AI has considerable potential to enhance inspection processes, its implementation requires appropriate validation and human oversight. Ultimately, integrating AI into pharmaceutical quality control may contribute to improved manufacturing reliability and patient safety.
Can AI and Human Creativity Shape the Future of Pharmaceutical Innovation?
Rawan Heany
What if artificial intelligence could help scientists discover pharmaceutical formulations they might never have imagined? Artificial intelligence (AI) is emerging as a promising tool in pharmaceutical formulation, offering new ways to develop and optimize medicines. AI can analyze large amounts of data, identify hidden patterns, predict formulation properties, and suggest innovative ingredient combinations.
Beyond efficiency, AI introduces a fascinating possibility: contributing to scientific creativity by identifying solutions that human scientists might overlook. Could the collaboration between human intelligence and artificial intelligence transform pharmaceutical formulation?
This presentation explores AI's potential as a creative partner, examining its capabilities, limitations, and practical applications while highlighting the essential role of pharmaceutical scientists in ensuring safety, quality, and feasibility.
Ultimately, the future of pharmaceutical innovation may depend not on choosing between human creativity and artificial intelligence, but on discovering what they can achieve together.
Finding New Uses for Existing Medicines
Khadeja Mouaffaq Ballan
Drug repurposing, also known as drug repositioning, involves identifying new therapeutic uses for existing medicines. This approach has gained attention because traditional drug development is often time-consuming and expensive. Artificial intelligence (AI) can support drug repurposing by analyzing biological and medical data to identify potential relationships between existing drugs and diseases.
This review explores how AI can accelerate drug discovery by predicting new therapeutic applications and identifying promising drug candidates. Its potential applications include cancer, neurodegenerative disorders, and rare diseases. By utilizing existing medicines, this approach may reduce research time and development costs.
However, AI predictions alone cannot guarantee drug safety or effectiveness. Laboratory experiments, clinical trials, and regulatory assessments remain essential to validate potential treatments. Challenges such as limited datasets, poor data quality, and difficulties interpreting AI predictions may also affect reliability.
In conclusion, combining AI with experimental research could improve drug discovery efficiency and support the development of innovative therapeutic options.
Ithar Younus
Language within educational settings transcends its traditional role as a mere medium for transmitting information; it serves as a fundamental cognitive architect and a primary determinant of students’ psychological well-being. Grounded in psycholinguistics and educational psychology, this study investigates how instructional language structures, teacher discourse, and linguistic diversity influence learners' cognitive processing and emotional states. Drawing upon Vygotsky’s Social Constructivism and Sweller’s Cognitive Load Theory, the research highlights how linguistic complexity in curriculum delivery can induce extraneous cognitive load, thereby hindering working memory capacity and information retention. Furthermore, the psychological dimensions of classroom language are examined through the lenses of Krashen’s Affective Filter Hypothesis and Dweck’s Growth Mindset framework. Empirical evidence indicates that evaluative and rigid linguistic environments trigger Foreign Language Anxiety (FLA) in approximately 30% to 40% of learners, creating an affective barrier to knowledge acquisition, whereas effort-oriented, supportive teacher discourse significantly fosters psychological safety, academic resilience, and intrinsic motivation. In alignment with contemporary technological shifts, this study also explores the emerging role of Artificial Intelligence (AI) in reshaping educational linguistics. Recent findings suggest that AI-driven language tools provide personalized linguistic scaffolding and non-judgmental interactive spaces, which effectively mitigate language anxiety and optimize cognitive load. Ultimately, this paper proposes an integrative pedagogical framework that combines psycholinguistically informed teaching practices with AI-assisted tools to cultivate inclusive, cognitively efficient, and psychologically empowering educational environments.
Evidence from Syrian Humanitarian Organizations
Ahmad Alyousef1; Amjad Alhussein2; Bashar AL Agha3
Humanitarian organisations encounter significant logistical and technical obstacles in sustaining temperature-sensitive medical cold chains, including those for vaccinations, pharmaceuticals, and blood supplies. The extensive implementation of digital transformation has transformed humanitarian logistics by providing novel solutions to address cold chain challenges; however, empirical evidence on how these technologies improve humanitarian cold chain performance, especially in the unpredictable circumstances of humanitarian operations, remains scarce. This study uses the Dynamic Capabilities View (DCV) to analyse the influence of contemporary technologies, namely the Internet of Things (IoT), blockchain technology (BC), and artificial intelligence (AI), on the improvement of humanitarian cold chain performance (HCCP), while accounting for the mediating effects of supply chain collaboration (SCC) and supply chain agility (SCA). Data were collected via a standardised questionnaire disseminated to humanitarian and supply chain organisations functioning in Syria. Out of 272 received questionnaires, a total of 257 valid responses were retained after removing incomplete blank submissions and analysed via Partial Least Squares Structural Equation Modelling (PLS-SEM) with 5,000 bootstrap subsamples. The results demonstrate that AI (β = 0.280, p < 0.001; β = 0.233, p = 0.001), IoT (β = 0.293, p < 0.001; β = 0.176, p = 0.015), and blockchain technology (β = 0.217, p = 0.004; β = 0.214, p = 0.010) positively and significantly influence both SCC and SCA. Moreover, both SCC (β = 0.340, p < 0.001) and SCA (β = 0.436, p < 0.001) positively affect HCCP, with SCA exhibiting the strongest direct impact on performance (R² = 48.3%). Mediation analysis confirms that organisational capabilities (SCC and SCA) play a significant mediating role in translating digital transformation technologies into improved humanitarian cold chain outcomes across all six indirect pathways. This study contributes to the humanitarian logistics and supply chain management literature by extending the application of the Dynamic Capabilities View to humanitarian cold chain operations in conflict-affected environments. It also provides empirical evidence from Syria, a context that remains underrepresented in the existing literature. The findings offer practical implications for humanitarian organisations seeking to strengthen cold chain performance through investments in digital technologies.