Undeclared AI use as academic misconduct

Question
Can failing to declare or acknowledge permitted AI use be academic misconduct?

Short answer

Yes. In England and Wales, failing to declare or acknowledge AI use can amount to academic misconduct even where the AI use itself was permitted. The misconduct in that situation is not the use of the tool but the failure to be transparent about it, which breaches the assessment rules and the requirement that submitted work is honestly attributed. Whether it will be treated as misconduct in any particular case depends on the wording of the institution’s own regulations and the specific assessment brief, so those documents are decisive.

Why non-declaration can itself be misconduct

Academic misconduct in higher education is generally defined as gaining, or attempting to gain, an unfair advantage in assessment, or misrepresenting the origin or authorship of work. Most university regulations frame it in terms of academic integrity and honest attribution rather than a closed list of banned acts.

Where an institution permits AI use but attaches a condition that the use be declared, cited, referenced or reflected upon, that declaration requirement forms part of the assessment rules. Breaching a compulsory condition of the assessment is capable of being misconduct in its own right, for two reasons. First, presenting AI-assisted work as if it were entirely your own unaided work misrepresents authorship. Second, ignoring a mandatory instruction in the brief can be treated as a breach of the assessment regulations regardless of whether an advantage was actually gained.

University guidance in England reflects this. Institutional policies commonly state that a student who fails to acknowledge AI use, or who refuses to declare, cite, reference or reflect on it, may be treated as attempting to hide that the work is not their own, and will be referred to an academic misconduct process. That is why non-disclosure of otherwise permitted use is routinely capable of being caught.

The key distinction: permitted use versus disclosed use

It is important to separate two questions that are often confused.

Whether the AI use was allowed at all. This depends on the assessment brief and module or programme rules. Some assessments permit AI freely, some permit it only for limited purposes such as proofreading or idea generation, and some prohibit it entirely.

Whether the use was properly declared. Even where use is allowed, many institutions require an acknowledgement, an AI-use statement, or a citation and reference.

Misconduct can arise from failing either test. A student can use AI in a way that was fully permitted in substance, yet still commit misconduct by not declaring it where declaration was required. Conversely, some institutions make clear that you will not be penalised for declaring use that was in fact permitted, so honest declaration is generally protective. Note, though, that a full and honest declaration does not cure use that went beyond what the assessment allowed. If the underlying use was not permitted, disclosing it does not make it permissible, although it will usually be treated far more leniently than concealment.

Factors that affect whether it will be treated as misconduct

Several factors influence the outcome, and they matter because materially different facts produce materially different results.

The precise wording of the regulations. If declaration is expressly mandatory, non-declaration is a clear breach. If the guidance is vague, aspirational or merely encouraged rather than required, a student has a stronger argument that no rule was actually broken.

Whether the requirement was clearly communicated. Fairness and natural justice principles require that students be told what is expected. If the declaration requirement was buried, ambiguous, changed mid-course, or not brought to students’ attention, that undermines a misconduct finding and is a legitimate ground of challenge.

Intent and honesty. Many panels distinguish between deliberate concealment and an innocent or careless omission. A genuine, reasonable misunderstanding about whether a declaration was needed is often treated differently from a deliberate attempt to pass off AI output as one’s own. Some regulations require intent to gain unfair advantage; others treat misconduct as capable of arising without intent but treat lack of intent as strong mitigation on penalty.

The nature and extent of the undeclared use. Undeclared use of AI to draft substantial portions of the work is far more serious than an undeclared grammar check. The greater the reliance, the more the non-declaration looks like misrepresentation of authorship.

Whether an unfair advantage was actually obtained. Where a real advantage was gained, the case for misconduct is stronger. Where the undeclared use was trivial and gave no advantage, that is a point in mitigation.

Evidence and how these cases are proved

If you are facing an allegation, understand that the institution normally bears the burden of proof, usually on the balance of probabilities. AI detection tools are not reliable enough to prove authorship on their own and are widely criticised, so an allegation resting solely on a detector score is vulnerable to challenge. Institutions often support allegations with other evidence such as inconsistencies in style, an inability to explain the work at a viva or authenticity meeting, metadata, or version history.

If you are the student, the most useful evidence you can preserve is your working process: drafts, notes, browser or document version history, and records of exactly how and when you used any AI tool. This helps show either that your use fell within what was permitted or that any failure to declare was innocent.

Practical next steps if you are accused

1. Obtain and read the exact regulations. Get the academic misconduct policy, the AI or academic integrity policy, and the specific assessment brief that applied. The wording of these documents will usually decide the case.

2. Identify precisely what rule is said to have been broken. Distinguish between an allegation that your use was not permitted and an allegation that you failed to declare permitted use. The defences differ.

3. Check whether the declaration requirement was mandatory and clearly communicated. If it was unclear, optional, or not properly notified, say so.

4. Prepare your account and evidence. Explain honestly what you did, produce drafts and version history, and be ready to demonstrate your understanding of the work at any viva.

5. Engage with the process and use support. Most institutions allow you to be accompanied by a students’ union adviser at an academic misconduct hearing, and the students’ union advice service is usually free and experienced in these cases.

6. Use the appeal route if the finding is wrong or the penalty is disproportionate. After exhausting internal procedures you can, in most cases, ask the institution for a Completion of Procedures letter and refer the matter to the Office of the Independent Adjudicator for Higher Education, which reviews procedural fairness and reasonableness rather than remaking the academic judgement.

What would sharpen this advice

The answer turns heavily on facts I do not have. It would help to know which institution and level of study is involved, the exact wording of the assessment brief on AI, whether the policy makes declaration mandatory or merely recommended, how you actually used AI, and what evidence the institution is relying on. If you can share the relevant policy wording and the assessment brief, the position can be assessed much more precisely, because these documents, not general principle, will ultimately determine whether undeclared permitted use is treated as misconduct in your case.

Current sources checked

This answer draws on broad legal knowledge and checks current law, guidance and procedure against relevant sources.

University Guidance on the use of Generative Artificial Intelligence by students and staff, in learning, teaching, and assessmentliverpool.ac.ukUse of Generative AI tools | Student Handbook | Loughborough Universitylboro.ac.ukGenerative AI and academic integrity | StudySkills@Sheffield | The University of Sheffieldsheffield.ac.ukArtificial Intelligence and Academic Integritywarwick.ac.uk
Verify important information before relying on it.
Was this helpful?
0 people found this helpful

Ready to stop guessing?

The above is just an example answer. Sign up now to get personalised guidance and ask follow-up questions based on your own situation. Bring your legal issue, your contract, or your question. Lawyer Destroyer gives you clear, practical guidance so you can move forward with confidence.