Accused of AI plagiarism at university: what happens next
An allegation is the start of a process, not a finding. Here is how that process usually runs, what a detection score can and cannot show, and how to prepare.
If your university suspects AI plagiarism, you will normally receive a formal message describing the concern and inviting you to explain your work. Nothing has been decided at that point. Read the message carefully, note the deadline for replying, and start gathering the drafts and notes that show how the assignment was written.
Most allegations are resolved through a meeting and a written decision, and a clear, well evidenced response makes a real difference to the outcome. This guide covers each stage in turn, for UK universities.
What an AI plagiarism allegation means
UK universities treat undeclared use of generative AI as academic misconduct when an assessment did not allow it. The rules differ between institutions and often between modules, so the first document to find is your assessment brief. Many now state whether AI tools were banned, permitted for limited tasks such as planning, or allowed with a declaration.
That brief matters because the question in front of the panel is whether you broke the rules for that assessment, not whether AI was involved anywhere. Using a spellchecker where spellcheckers were allowed is not misconduct. Pasting generated paragraphs into an essay that banned AI is.
Your university's academic integrity regulations set out the process, the possible outcomes and your rights. They are usually public and worth reading in full before you reply to anything.
How Turnitin AI detection works
Turnitin AI detection gives a percentage estimate of how much of a submission its model predicts was generated by AI, with highlights on the passages concerned. It is a probability, not a record of how the text was produced. Turnitin shows scores between 1 and 19 percent as an asterisk with no highlights, because its own testing found more false positives in that range.
Detection tools are also known to misread some kinds of genuine writing. Research published in 2023 by Liang and colleagues found that several detectors flagged essays by non-native English writers as AI generated far more often than essays by native speakers. Formulaic academic prose and text heavily reworked by grammar tools can raise scores too.
None of this means a score can be ignored. It means the score starts a conversation, and the rest of the evidence decides it.
How the process usually runs
The notification
The first message should explain what is alleged and on what basis. If it does not include the evidence, ask for it in writing. You are entitled to know what you are responding to.
The meeting
Many universities hold an informal or formal meeting with an academic integrity officer or panel. Expect questions about the work itself: how you found sources, why the argument runs in a certain order, what a particular sentence means. Students who wrote their own work can usually answer these in detail, and that knowledge is persuasive.
The outcome
Outcomes range from no case to answer, through a warning or a reduced mark, to a capped resubmission. More serious penalties are generally kept for serious or repeated misconduct. Your regulations will list the options and the factors that move a case between them.
Collect evidence that the work is yours
The strongest evidence shows the writing developing over time. Version history in Word, OneDrive or Google Docs records when text was added and changed. Early drafts, plans, handwritten notes, reading notes with page numbers, and emails to your supervisor all show a process that generated text does not leave behind.
Print or save copies before the meeting, and annotate your submission with the sources behind each section. If a detector highlighted particular paragraphs, be ready to talk through how you wrote those passages specifically.
Prepare for the meeting properly
Contact your students union advice service as soon as the allegation arrives. Advisers know the local regulations, can review your response, and can often attend with you. It costs nothing and it is independent of your department.
Be precise and honest about any tools you used. If you ran a draft through a grammar tool or asked a chatbot to explain a concept, say so plainly and explain what the brief allowed. Panels respond far better to an accurate account than to a denial that later collapses.
Resist the urge to rewrite the work or run it through a so called humanising tool after the allegation arrives. Changing the evidence looks exactly like what it is, and it turns a question into a problem.
If the outcome feels wrong
Universities allow appeals, usually on defined grounds such as a procedural error or new evidence that was not available at the time. Deadlines are short, often two or three weeks, so check them the day the decision arrives.
Once the internal process is finished, you will receive a Completion of Procedures letter. In England and Wales the Office of the Independent Adjudicator can then review the case. Students in Scotland complain to the Scottish Public Services Ombudsman, and students in Northern Ireland to the Northern Ireland Public Services Ombudsman.
Use AI without risking an allegation
Check what each assessment permits before you open any tool, and declare any use the brief asks you to declare. Keep your drafts and version history switched on for every piece of work, not just the ones that feel risky. Write in your own voice from your own notes, and when you draw on a source, put its ideas into your own words and cite it in your department's style, which for most UK courses isHarvard referencing. Our guide on how to paraphrase covers that skill in detail.
Our academic integrity policy explains where we draw the line on the support we offer, and the guide touniversity proofreading policies explains what outside help is allowed.
If an AI plagiarism allegation has shaken your confidence in your writing, one to one coaching can help you build a clear academic voice for the resubmission and the work after it, with every word still written by you.
Last updated: 14 September 2026 · Written by the AssignWise editorial team ·Spotted something wrong?
Frequently asked questions
No. The AI writing indicator is a prediction about how likely text is to be machine generated, not proof of how it was written. Turnitin itself hides scores between 1 and 19 percent behind an asterisk because false positives are more common in that range. Universities are expected to weigh the score alongside other evidence.
Usually a member of staff explains the concern and asks you about the work: how you researched it, why you made particular choices, and what some passages mean. You can normally bring a supporter from your students union. The decision tends to come afterwards in writing, not in the room.
Describe exactly what you used and how, and check what your assessment permitted. Many universities allow spelling and grammar tools but not tools that generate or rewrite content. An accurate account early on is almost always treated better than a denial that later falls apart.
Severe outcomes exist for serious or repeated cases, but a first allegation more often ends in no further action, a warning, a reduced mark or a capped resubmission. Your university regulations list the penalties and the factors that decide between them, so read that section before any meeting.
Your students union advice service is free, independent of the academic department, and deals with these cases constantly. Your personal tutor can explain local practice. If the internal process ends and you still believe the outcome is wrong, the Office of the Independent Adjudicator reviews complaints about universities in England and Wales.
Want a second pair of eyes on it?
Send us your document for a fixed, no-obligation quote. We reply within one working hour.
Get Your Free Quote