Understanding ethical dimensions of AI in higher education: insights from faculty members and students

Synthesis: Bilgiç and Sever used an explanatory sequential mixed-methods design to examine faculty and student views on the ethical use of AI in higher education, surveying 971 students and 135 faculty members followed by semi-structured interviews with 23 students and 14 faculty. Both groups expressed supportive views toward ethical AI use, but in different forms across individual, technological, institutional, and societal levels: faculty emphasized ethical principles while flagging a lack of institutional guidelines, whereas students valued AI's learning benefits but voiced uncertainty about the sharing of ethical responsibilities. Qualitative analysis surfaced six themes exposing a multidimensional ethical structure, with shared concern that excessive AI use could weaken cognitive skills. The authors call for faculty professional development, ethics courses in curricula, clear institutional guidelines, and interdisciplinary collaboration.

Core Finding

Faculty members and students hold supportive yet distinctly shaped views on AI ethics in higher education, with faculty focused on principles and institutional support gaps and students on learning benefits and uncertain ethical responsibility. Both groups pragmatically accept AI while aware of its risks, and both worry that overuse erodes cognitive skills — a shared concern that frames ethical AI integration not merely as technology adoption but as preserving students' cognitive and professional development.

Faculty Perspectives

Faculty survey responses scored highest on the Equality, inclusiveness, and justice dimension (M=4.10) and lowest on Institutional support (M=3.81). Notably, faculty strongly supported institutional measures to prevent unethical AI use (M=4.59) while rating the adequacy of institutional guidelines lowest (M=2.99), revealing that a high level of individual ethical responsibility does not always align with sufficient structural support. Faculty expressed limited knowledge of data privacy and cultural inclusivity in AI-supported materials, and some were hesitant to intervene in students' AI use out of respect for student autonomy — underscoring the need for institutional frameworks that translate individual ethical awareness into a sustainable ethical culture.

Student Perspectives

Students scored highest on Ethical awareness and responsible use (M=3.95) and lowest on Academic integrity (M=3.72). Their top item was recognizing AI results as recommendations with ultimate responsibility belonging to oneself (M=4.20), yet they were less likely to actively check whether tools had ethical guidelines. Students tended to interpret ethics through flexible, individual frameworks — contextual responsibility and subjective truths — attributing output ownership or errors to AI, a reflection of legal uncertainties that leave room for their own ethical reasoning. This is better read as a conflict between ethical paradigms than a simple ethical deficiency, pointing to the need for clear institutional policies.

A Multidimensional Ethical Structure

Qualitative analysis revealed six themes spanning data ethics, algorithm ethics, and pedagogical ethics, consistent with the framework in which AI ethics in education forms an interconnected causal chain from fundamental principles (transparency, accountability, fairness, autonomy) to behavioral outcomes. Both groups shared concerns that AI could weaken cognitive skills long-term, that curricula must be updated, and that professional-life uncertainties have arisen. The bottom-up risk-management process that emerged — faculty sharing usage criteria, students verifying content — falls short without an institutional roadmap, and the contrast between faculty's perception of students' "ethical inadequacy" and students' "acceptance of ethical responsibility" complicates governance. Concerns about overuse strengthening superficial learning habits and weakening critical thinking connect directly to cognitive offloading and AI-misuse harms.

Relevance to the knowledge base

This paper grounds the knowledge base's treatment of Ethics in Higher Education with a dual-stakeholder, mixed-methods account of how faculty and students actually reason about AI, bridging AI in Education with AI Governance and institutional policy. Its emphasis on faculty professional development, curriculum reform, and clear guidelines offers concrete levers for educators and administrators. The documented risk that excessive AI use may weaken Critical Thinking and cognitive skills connects the paper to the knowledge base's coverage of Cognitive Offloading and AI Misuse and Learning Harm, while its privacy and academic-integrity themes align with Privacy and Academic Integrity concerns.

What this means for practice

• Instructors. Treat students' stated ethical responsibility as a starting point rather than a guarantee: students scored highest on ethical awareness (M=3.95) but lowest on Academic Integrity (M=3.72), and few checked whether tools had ethical guidelines.
• Build explicit data-privacy and cultural-inclusivity content into AI professional development, where faculty self-reported their weakest knowledge.
• Pair any institutional guideline with classroom-level Ethics instruction, since faculty rated the need for institutional measures highly (M=4.59) but the adequacy of current guidelines lowest (M=2.99).
• Require students to document how they verified a tool's guidance, countering their tendency to attribute output ownership or errors to the AI.

Limitations

• Data come from a single public university in Turkey, and the authors note the study is situated in a specific context that reflects participants' reported views rather than observed practice.
• Both phases rely on self-report, a survey of 971 students and 135 faculty followed by interviews with 23 students and 14 faculty, which introduces subjective interpretation.
• Only 135 of 776 faculty responded to the survey, and the authors flag potential non-response bias among faculty who chose not to participate.
• The sequential design is cross-sectional, so it cannot track how ethical perceptions and practices change over time.

Citation

Bilgiç, B., & Sever, D. (2026). Understanding ethical dimensions of AI in higher education: insights from faculty members and students. International Journal of Educational Technology in Higher Education.