How Teachers Should Use a Turnitin AI Score

Turnitin has said this in more than one place. In its FAQ on AI writing detection: "Hence, we must emphasize that the percentage on the AI writing indicator should not be used as the sole basis for action or a definitive grading measure by instructors."

A second official page says the same thing in different words: the model "may not always be accurate," so it "should not be used as the sole basis for adverse actions against a student," and determining misconduct "takes further scrutiny and human judgment."

The message is consistent across Turnitin's own materials: the number is not the decision. The decision belongs to the educator, and the vendor writes separately about what an instructor is told to do when the AI percentage is high.

Turnitin's guidance for reviewing the report puts it directly: the AI score "is not meant to provide definitive answers in isolation."

The same page explains why: "More important than any tool is the educator who sees the score and makes decisions balancing this information with their personal knowledge of their students, their work, and institutional policy."

And it closes with the intended use: "When educators look at the AI writing score and utilize it as a single data point rather than a definitive response, then it is being used as intended." That sentence is the whole job description for how to read the score.

Turnitin also lists the kinds of text that are prone to false positives — human writing flagged as AI: "content without a lot of structural variation, text that literally repeats itself, or text that has been paraphrased without developing new ideas."

The follow-up is the part to remember: "If our indicator shows a higher amount of AI writing in such text, we advise you to take that into consideration when looking at the percentage indicated."

In practice this means a student essay that stays flat in structure, reuses its own wording, or paraphrases a source without adding new ideas can carry a higher score without the text actually being AI-generated. Turnitin itself advises discounting such scores. It is also why paraphrasing can make a Turnitin AI score go up rather than down.

Start from the official baseline: the score "should not be used as the sole basis for action." A percentage alone is not evidence of misconduct.

Turnitin's own framing is that more text makes predictions more reliable, and that short documents are the weakest case. Its FAQ notes that on documents of only a few hundred words, "the prediction will be mostly 'all or nothing'." So on short submissions, treat the number with extra caution.

When a score looks high, look at the text, not just the number: is the writing flat, repetitive, or paraphrased without new ideas? If yes, that is exactly the pattern Turnitin says to weigh differently; how essay structure affects a Turnitin AI score covers the first of those three. When a score looks low, remember the 1% to 19% range is displayed as an asterisk rather than a number (what the asterisk (*%) means on a Turnitin AI score) — no percentage means no strong signal, not a clean bill of health.

For a student, the practical takeaway is that a score is accompanied by the educator's judgment, not a verdict. Turnitin's release notes repeat the same line: an AI writing score "should not be used as the sole basis for adverse actions against a student."

If your instructor raises the score, the productive conversation is about the writing itself — structure, repetition, paraphrase without new ideas — because those are the features Turnitin names as prone to false positives. The score is a signal to look, not a statement that misconduct happened. For the student's side of that conversation, see flagged, but you wrote it yourself: preparing a response.