Beyond plagiarism: Why universities need AI governance, not just misconduct rules

When artificial intelligence (AI) first entered university classrooms, the ethical debate was framed in familiar terms: Could students submit AI-generated work as their own? Should using ChatGPT in an essay count as misconduct? How could instructors tell the difference between legitimate assistance and unacceptable substitution? Those questions still matter but are no longer enough. As AI spreads across higher education, it is no longer confined to essay writing or take-home assignments. It is shaping teaching, assessment, thesis supervision, research design, data handling, and even administrative decision-making. AI ethics is no longer only about misconduct, it is about governance.

The misconduct lens is too narrow

The problem with a plagiarism-centered view of AI is not that it is wrong. It is that it is too limited for the scale of change now underway. If universities treat AI only as a new way for students to cheat, they reduce a complex institutional challenge to a disciplinary one. That may produce tougher rules for assignments, but it does little to answer broader questions such as: Can generative AI be used in thesis writing? Should researchers be allowed to rely on it in grant proposals or literature reviews? What happens when staff upload sensitive data into external platforms? Who is responsible when AI-generated outputs are biased, inaccurate, or fabricated?

AI is reshaping the whole university

What makes AI different from earlier academic and research integrity challenges is its reach. It does not affect only one part of university life: in teaching, it changes how students write, study, and interact with learning materials. In assessment, it challenges old assumptions about what independent work looks like. In research, it raises difficult questions about disclosure, authorship, confidentiality, and the reliability of machine-generated outputs. In administration, it creates new concerns around procurement, automation, risk management, and the use of institutional data.

What governance looks like in practice

Once AI touches all of these domains, ethics can no longer be handled through a single misconduct code or a paragraph in an academic integrity policy. Universities need a broader framework that links academic values to operational practice. Responsible AI governance requires institutions to decide who is accountable, what must be disclosed, which uses are acceptable, and where the boundaries lie. That may include rules requiring students and researchers to declare how AI was used in producing work. It may require guidance for instructors on designing assessments in an AI-rich environment; policies on whether sensitive data can be entered into commercial AI systems; ethics review procedures for research projects that use AI in ways that affect human participants, privacy, or public trust.

Governance also means building the structures that allow these rules to function. Universities need leadership, training, coordination, and review mechanisms. They need clarity over who makes decisions and who monitors implementation. They also need policies that connect teaching, research, and administration rather than treating them as separate worlds.

Different systems, same lesson

In some contexts, institutions are moving toward comprehensive AI frameworks that cover coursework, examinations, thesis writing, research, and data protection. In others, universities are developing guidance through task forces, committees, and funder requirements, often in more decentralized ways. Elsewhere, national ambition around AI is advancing faster than institutional capacity, leaving some universities much better prepared than others.

Despite these differences, the pattern is clear: where universities treat AI as a whole-of-institution issue, the focus shifts from simple prohibition to questions of disclosure, oversight, review, and accountability, offering a workable long-term response to a technology that is already embedded in academic life. 

Why capacity matters

Not all universities have the same ability to govern AI well. Some institutions have the resources to establish committees, draft detailed guidance, train staff, review risks, and adapt policies across multiple domains. Others do not. For less-resourced institutions, responsible AI use may be expected in principle but difficult to implement in practice.

That makes AI ethics not only a matter of rules, but of equity and capacity. A university cannot ensure responsible AI use if it lacks staff expertise, governance systems, or data protection safeguards. It cannot build fair and transparent oversight if responsibility is unclear or scattered. And if only the best-resourced universities can develop robust governance, then AI may deepen the gap between institutions rather than support the higher education sector as a whole.

From suspicion to responsibility

Many early university responses to AI were driven by suspicion: how to detect it, restrict it, or punish its misuse. That reaction was understandable. But it is not enough for the phase higher education is entering now. Universities need to move from suspicion to responsibility.

That means asking not only whether AI has been used, but how, by whom, for what purpose, and under what safeguards. It means creating systems in which disclosure is meaningful, oversight is credible, and accountability does not disappear behind the machine. It means connecting academic integrity to research ethics, data governance, and institutional leadership, accepting that AI ethics is now part of the core governance of higher education.

A broader ethical challenge

In the era of postplagiarism, the real question is no longer whether AI creates new opportunities for misconduct. The deeper issue is whether universities can govern AI in ways that protect trust, uphold academic values, and distribute responsibility fairly across the institution. AI ethics in higher education, therefore, is about far more than plagiarism. It is about how institutions manage technological change, preserve human judgment, and ensure that innovation does not outpace accountability.

Universities that understand this will be better prepared for the future. Those that continue to treat AI merely as a cheating problem may find they have misread the challenge altogether.

Acknowledgment

This project was supported by the Swiss foundation Movetia. The sponsor influenced neither the research design nor the interpretation of the results.

References

• Denisova-Schmidt, E., Altbach, P., & de Wit, H. (Eds.) (2025). Handbook on corruption in higher education. Edward Elgar Publishing. 
• Eaton, S.E. (2023). Postplagiarism: transdisciplinary ethics and integrity in the age of artificial intelligence and neurotechnology. International Journal of Educational Integrity, 19, 23.
• Kuzhabekova,A., Kim, T., & Mamyrbekov. A. (2026). Factors of research misconduct in transitional contexts: perceptions of faculty from Kazakhstan. Higher Education Quarterly, 80, no. 3: e70135.
• Xu, Z., & Denisova-Schmidt, E. (2026). The Chinese dilemma of research misconduct. Higher Education Quarterly, 80, no. 2: e70131.