Turnitin Flagged Your Work. Here's What's Happening.
Why Turnitin produces false positives
Turnitin's AI writing detection works by analysing the statistical patterns in finished text — things like sentence complexity, vocabulary distribution, and perplexity scores. The fundamental problem is that human writing that scores high on these metrics looks identical to AI-generated text by the same measure.
Three groups are disproportionately affected: non-native English speakers (who tend toward formal constructions), writers in formal genres (academic, legal, technical), and careful editors (writers who revise extensively toward clarity). All three produce text that resembles AI output by statistical measure.
Turnitin's detection reports a percentage — not a verdict. A score of 80% does not mean 80% of the text was AI-written. It means Turnitin assessed 80% of the sentences as having statistical patterns associated with AI generation. A careful human writer in an academic genre can easily produce text that scores 90%+.
What Turnitin's own documentation says
Turnitin has published guidance explicitly stating that AI writing indicators should not be used as the sole basis for academic misconduct decisions. Their official position:
“Turnitin's AI writing detection capability is designed to help educators identify text that may have been generated by AI, and should not be used as the sole basis for academic misconduct decisions.”
This is significant. Turnitin's own terms of use acknowledge that the indicator is a signal for investigation, not a finding of misconduct. Any institution using Turnitin's AI score as a standalone verdict is misusing the product per Turnitin's own guidance.
Use this in your appeal. If your instructor is treating the Turnitin score as conclusive proof, you can cite Turnitin's own guidance to the contrary.
Evidence you likely already have
Google Docs version history
Strongest evidenceEvery edit with timestamps. Shows your document evolving from blank to final over hours or days. Access via File → Version history → See version history.
Draft files and autosaves
Strong evidenceMultiple saved versions showing the document's evolution. Check OneDrive, Google Drive, or your local Downloads folder.
Research notes and browser history
Supporting evidenceAI doesn't leave a search trail. Your browser history and notes show the intellectual work that preceded the writing.
File timestamps
Supporting evidenceRight-click any file → Properties (Windows) or Get Info (Mac) to see when it was created and modified. Screenshot these.
Frequently asked questions
Can I dispute a Turnitin AI score?
Yes. The appeal process is the same as any academic integrity concern — you present your evidence to the instructor or academic integrity office. Turnitin's own guidance says their score should not be used as standalone proof, which you can cite explicitly in your appeal.
Does Turnitin's score affect my grade automatically?
No. Turnitin flags potential issues for instructor review — it doesn't integrate directly with grading systems. The instructor makes the decision. An appeal with evidence goes to the instructor or the institution's academic integrity process, not to Turnitin.
What if I don't have Google Docs history?
Any evidence of process helps: file timestamps, email drafts, notes, research history. If you have none of this, your appeal will rest more on the statistical nature of Turnitin's method and on Turnitin's own guidance about false positives. See the /accused guide for a full evidence checklist.
Prevent this from happening again
Scripli records your writing session before you submit. If Turnitin flags your next essay, you'll have independent, verifiable proof — not just your word.