Part of the flagged-as-AI hub · Updated August 23, 2026
Does Turnitin detect AI?
Yes — Turnitin returns an AI-writing score. But it does not do what the question assumes. It does not catch anyone, identify any tool, or produce evidence. It estimates how closely your sentences resemble machine-written ones, and on careful academic English by a non-native speaker that estimate is wrong far more often than anyone comfortable with it would like.
What does the Turnitin AI score actually measure?
Short answer: the AI indicator is a separate number from the similarity score, and it measures statistical texture rather than sources — the share of the document a model considers likely machine-generated. Nothing in it identifies a tool, a prompt or a source.
Turnitin runs two separate checks, and conflating them causes more panic than either deserves.
| Similarity score | AI indicator | |
|---|---|---|
| How long it has existed | Decades | Newer, and independent of the similarity check |
| What it measures | Overlap with sources Turnitin has seen before | The share of the document a model considers likely machine-generated |
| What it compares your text against | Existing sources | Statistical texture — nothing about sources at all |
A 20% similarity score and a 20% AI score mean nothing alike — see similarity score for why the two numbers on your report are not comparable.
The AI indicator works on statistical texture: how predictable each next word is (perplexity) and how much sentence rhythm varies (burstiness). Machine text tends to be smooth and even. So does careful formal writing. The detector cannot tell the difference between «this was generated» and «this was written by someone being very careful», because at the level it measures, there is no difference to see.
Nothing in the output identifies a model. A passage from ChatGPT, from Claude, and from a diligent student who revised until the prose was even all land in the same place.
How often is Turnitin's AI detection wrong — and on whom?
Short answer: a Stanford study found AI detectors falsely flagged 61% of TOEFL essays written by non-native English speakers, while classifying essays by US-born students almost perfectly (Liang et al., Patterns, Cell Press, 2023).
The reason is mechanical rather than malicious. English learned through instruction is built from studied constructions and a deliberately chosen vocabulary. It is more uniform than English absorbed by ear — which is exactly the property the detector scores as machine-like. Writing produced across two languages carries the same signature; we call it translationese.
Institutions and vendors have drawn their own conclusions from detector false positives:
- Vanderbilt University disabled Turnitin's AI detection entirely in August 2023, citing false positives and the vendor's inability to explain individual results.
- Independent testing across fourteen detection tools concluded they are «neither accurate nor reliable» (Weber-Wulff et al., International Journal for Educational Integrity, 2023).
- OpenAI withdrew its own AI-text classifier in July 2023, six months after launching it, citing low accuracy.
More in how common false positives are and how accurate Turnitin's detection is.
Worried Turnitin will flag an essay you actually wrote?
Short answer: then you are the reader this page was written for — most people who arrive here wrote the work themselves and are afraid the number will say otherwise.
What this page will not tell you is how to get around the detector. Not out of primness — because the advice is worthless. Text rewritten to defeat a detector reads worse, and synonym-spinning produces the «tortured phrases» that have caused journal retractions on their own. If you did use a model to write for you, no page on this site will help you hide it.
What actually protects you if Turnitin flags your work?
Short answer: not a counter-score from a second detector, but evidence that the document grew — version history, earlier drafts carrying your own corrections, outlines, notes.
A committee has no reason to prefer one detector's estimate over another's, which is why a second score rarely settles anything. What persuades is the ordinary debris of real work: the drafts, the outlines, the notes. If you have already been flagged, the flagged-as-AI playbook is the step-by-step, including how to appeal.
If you have not been accused yet, this is the cheap moment. Write where history is kept, keep your drafts, and work through the authorship checklist. For work where an accusation would be expensive, the Diglot Authorship Certificate records the writing process as an append-only, cryptographically signed log — proof that exists before anyone asks for it, rather than a defence assembled afterwards.
Turnitin and AI detection — questions
Does Turnitin detect AI?
It reports an estimate. Turnitin runs a separate AI-writing indicator alongside its similarity check, and returns a percentage of the document its model considers likely AI-generated. That number is a prediction about statistical patterns, not a detection of an act: it names no tool, no prompt, and no source. Turnitin itself instructs institutions to treat the score as a starting point for a conversation rather than as evidence.
Does Turnitin detect ChatGPT specifically?
No detector can identify which model produced a passage. Turnitin's indicator is trained to recognise text that looks statistically machine-like in general — not the signature of one product. Output from ChatGPT, Claude, Gemini or a human writing very evenly all land in the same bucket, which is precisely why the score cannot tell you who wrote something.
Is the AI score the same as the similarity score?
No, and confusing them causes a great deal of unnecessary panic. The similarity score measures overlap with existing sources and has existed for decades. The AI indicator is newer, separate, and measures something entirely different. A 20% similarity score and a 20% AI score mean nothing alike — see similarity score for the distinction.
How accurate is it for non-native English writers?
Worse than for anyone else, and this is documented. A Stanford study found AI detectors falsely flagged 61% of TOEFL essays written by non-native English speakers, while classifying essays by US-born students almost perfectly. The mechanism is not bias in the ordinary sense: detectors score statistical uniformity, and English learned through formal study is more uniform than English absorbed at home.
Should I check my essay with a detector before I submit?
You can, but hold the result loosely. Detectors disagree with each other, scores shift between drafts of the same document, and passing one tool tells you nothing about the one your institution runs. A score that reassures you today is not evidence you can show a committee tomorrow. Process evidence is.
What actually protects me if I am accused?
Evidence that the document grew over time: version history, earlier drafts with your own corrections, outlines, notes. Committees find a documented writing process far more persuasive than a counter-score from a different detector. If the stakes are high, the Authorship Certificate records that process cryptographically as you write.
Diglot is a writing workspace for people who think in one language and publish in English — with a record of how the work was written. start for free, no credit card required.
