Synopsis
This workshop presents empirical findings from UAE-based focus groups exploring how students negotiate trust, responsibility, and reliance in everyday GenAI-supported learning. Moving beyond compliance and policy debates, it foregrounds students’ moral reasoning, verification practices, and conditional trust in tools such as ChatGPT. Using a Human Library methodology, student co-authors act as “human books” to facilitate structured dialogue on ethical boundary-setting and shared accountability. The session culminates in a co-design sprint to embed ethical reflection checkpoints into an Ethical Reflection Custom GPT, shifting institutional responses from enforcement toward scaffolded moral agency.
Abstract
The rapid normalisation of generative artificial intelligence (GenAI) tools such as ChatGPT has changed how students write, learn, and make academic decisions. Postplagiarism scholarship suggests that while hybrid human-AI writing is becoming the norm, responsibility for outputs remains with the human author (Eaton, 2023). In response, institutional approaches have largely focused on assessment redesign, detection tools, and policy development. While these address governance and compliance, they offer limited insight into how students themselves negotiate trust, verification, and moral responsibility in everyday GenAI use.
This workshop is situated within a broader research programme examining trust, agency, and moral responsibility in student use of GenAI (Denisova-Schmidt & Khan, 2026), including a UAE-based mixed-methods study (n = 567) that identified widespread adoption alongside persistent uncertainty regarding ethical use, permission, and institutional expectations (Khan et al., 2025). These findings revealed a disconnect between formal policy frameworks and students’ lived decision-making practices.
Building on this, the literature review indicates that student trust in GenAI is shaped by a combination of system-level factors (e.g., reliability and explainability), individual factors such as user confidence and expertise, and broader institutional and social contexts, including norms and expectations around acceptable use (Cui et al., 2025; Safarov et al., 2025; UNESCO, 2023). However, existing research provides limited insight into how students actively negotiate these factors in real-time academic contexts. This gap informed the design of the focus groups and, subsequently, the workshop itself.
Four online focus groups (n = 21), comprising school and university students, were conducted using a semi-structured discussion guide. Data were analysed using Reflexive Thematic Analysis (Braun & Clarke, 2006, 2022) to explore how students interpret trust, responsibility, and acceptable use in everyday GenAI-supported learning.
Findings indicated that students primarily positioned tools such as ChatGPT as learning support rather than authoritative sources, using them for summarisation, clarification, and self-testing. Trust was conditional rather than absolute, with students calibrating reliance based on their prior knowledge, trusting the tool more when they lacked understanding and relying on themselves when they felt confident in the topic. Participants reported awareness of inaccuracies, including fabricated references and incorrect information, prompting verification behaviours such as cross-checking against lecture materials and external sources. Responsibility was negotiated in relation to institutional expectations, perceived risk, and assessment stakes. Students indicated that permission-seeking was more likely in high-stakes tasks, while informal use was common in low-stakes contexts. Concerns about shared accountability emerged in collaborative work, where unverified AI-generated contributions could expose entire groups to academic risk. Participants also expressed concern about over-reliance, linking it to reduced critical thinking, diminished confidence, and potential long-term impacts on learning and professional competence (Zhai et al., 2024).
Building directly on these empirical and theoretical insights, this workshop translates research findings into an interactive, participant-led format designed to support ethical decision-making in GenAI-supported learning.
A key innovation of this workshop is the adaptation of the Human Library methodology to the context of GenAI and academic integrity. Traditionally used as a contact-based, experiential learning approach to promote dialogue and reduce bias, the Human Library is re-conceptualised here to foreground student moral reasoning in relation to AI use. Student co-authors act as “human books,” enabling participants to engage directly with lived experiences of negotiating trust, responsibility, and reliance in academic contexts.
Participants will engage in:
• Human Library dialogues with student “books” to explore lived experiences of GenAI use;
• facilitated small-group discussions analysing trust, responsibility, and verification practices; and
• a structured co-design sprint to develop ethical reflection prompts and decision checkpoints for integration into AI-supported learning tools.
Through these activities, participants will:
• gain insight into how students negotiate ethical decision-making in real academic contexts;
• critically examine assumptions about student responsibility and AI use;
• co-design practical strategies to embed ethical reflection into teaching and learning; and
• leave with adaptable artefacts and approaches applicable to their own institutional or classroom settings.
The workshop culminates in a co-design sprint where participants translate insights into embedded “moments of ethical reflection,” such as prompts, decision checkpoints, and risk signals that can be integrated into an Ethical Reflection Custom GPT (a prototype tool designed to prompt ethical decision-making during AI use). The session concludes with a facilitated synthesis discussion, with outputs captured through a shared digital platform to support transferability across contexts.
The workshop offers participants:
(1) empirically grounded insights into how students negotiate responsibility through contextual and relational factors rather than fixed rules;
(2) a novel application of the Human Library methodology for professional learning in AI ethics and academic integrity; and
(3) a co-design pathway for embedding ethical reflection into AI-supported learning environments.
This workshop is intended for higher education academics, educational developers, academic integrity practitioners, and AI ethics researchers, and is also relevant to doctoral students and school educators interested in trust, responsibility, and student agency in GenAI-supported learning.
Keywords: Generative AI, academic integrity, trust calibration, student moral responsibility, Human Library methodology
References
Braun, V., & Clarke, V. (2006). Using Thematic Analysis in Psychology. Qualitative Research in Psychology, 3(2), 77–101. https://doi.org/10.1191/1478088706qp063oa
Braun, V., & Clarke, V. (2022). Conceptual and design thinking for thematic analysis. Qualitative Psychology, 9(1), 3–26.
Cui, Y. L., Lin Zeng, M., Ke Du, X., & Miao He, W. (2025). What Shapes Learners’ Trust in AI? A Meta-Analytic Review of Its Antecedents and Consequences. IEEE Access, 13, 164008–164025. https://doi.org/10.1109/access.2025.3611367
Denisova-Schmidt, E., & Khan, Z. R. (2026). Designing for trust: Embedding moral responsibility in generative AI use in the classroom. Leading House MENA, University of Applied Sciences and Arts Western Switzerland (HES-SO). https://www.hes-so.ch/en/la-hes-so/international/leading-house-mena/projects/detail-projet/designing-for-trust-embedding-moral-responsibility-in-generative-ai-use-in-the-classroom-1
Eaton, S. E. (2023). Postplagiarism: transdisciplinary ethics and integrity in the age of artificial intelligence and neurotechnology. International Journal for Educational Integrity, 19(1). https://doi.org/10.1007/s40979-023-00144-1
Khan, Z. R., Aljanahi, M. H., Abutabenejh, S., Khan, A., Lal, R., & Merakeb, I. (2025). Student perceptions of GenAI use in the UAE. In European Conference on Ethics and Integrity in Academia (ECEIA25), Sweden: Book of Abstracts (pp. 128–130). European Network for Academic Integrity.
Safarov, C., Gadzinski, G., & Schlögl, S. (2025). In AI We Trust? Exploring the Role of Explainable GenAI and Expertise in Education. Human Factors: The Journal of the Human Factors and Ergonomics Society. https://doi.org/10.1177/00187208251401773
UNESCO. (2023, September 7). Guidance for generative AI in education and research. Unesco.org. https://www.unesco.org/en/articles/guidance-generative-ai-education-and-research
Zhai, C., Wibowo, S., & Li, L. D. (2024). The effects of over-reliance on AI dialogue systems on students’ cognitive abilities: a systematic review. Smart Learning Environments, 11(28), 1–37. https://doi.org/10.1186/s40561-024-00316-7