Direct answer
Turnitin uses AI writing detection alongside its traditional similarity (plagiarism) check to identify text that appears to have been generated by large language models such as ChatGPT, GPT-4, and similar tools. The "AI writing score" is a percentage in the AI writing report that shows how much of a submission Turnitin suspects was produced by AI, with a separate score for AI-written text that may have been paraphrased [1]. Any score below 20% is displayed as an asterisk (*%) rather than a single-digit percentage, so the only explicit low numeric outcome students typically see is 0% [1].
How Does Turnitin Detect AI Writing in Student Submissions?
Turnitin's AI writing detection was built specifically to recognize text produced by large language models, and it runs automatically when a paper is submitted through an enabled Turnitin account [2]. The tool does not rely on a single signal; it evaluates patterns across the whole document and then generates an AI writing report with an overall percentage plus a separate percentage for AI writing that may have been paraphrased or lightly edited [2].
Instructors see the result inside the Turnitin Feedback Studio document viewer, where sentences flagged as likely AI writing are highlighted so they can be reviewed in context rather than treated as a blanket verdict [2]. The indicator is designed as a complement to professional judgment, and Turnitin explicitly warns against using it as the sole basis for academic-integrity decisions [2]. Detection availability depends on institutional settings, license configuration, and language or region support, which is why not every account displays the indicator [1].
What Is the Difference Between the Turnitin AI Writing Score and the Similarity Score?
The similarity score measures how much of a paper matches text in Turnitin's databases of web pages, publications, and previously submitted student papers, while the AI writing score measures how much of the text appears to have been generated by an AI model — the two metrics are entirely independent [3]. A submission can have a very low similarity score yet a high AI writing score, or a high similarity score with a low AI writing score, because each indicator evaluates a different question about the document [3].
The similarity report shows match percentages per source and highlights overlapping passages, whereas the AI writing report shows an overall percentage, a paraphrased-AI-writing percentage, and highlighted segments suspected of being AI-generated [3]. Both reports are meant to be interpreted by instructors alongside assignment context and a conversation with the student, not as standalone proof of misconduct [3]. Students should keep the two numbers separate when reading their own reports, since confusing plagiarism with AI-generated text is a common misreading [3].
Can Students Preview Their Turnitin AI Writing Report Before Submitting?
Turnitin's guidance emphasizes that students should understand how the AI writing indicator works and how it is meant to be used before they ever see a score, so the percentage is interpreted fairly and calmly [4]. Because the detector is one piece of evidence among many, instructors are encouraged to discuss the report with students rather than treat the number as a final verdict on their work [4].
For students, the practical implication is straightforward: knowing how the indicator works ahead of time makes a flagged score much less alarming, especially since sub-20% results display only as an asterisk rather than a precise number [4]. If you want to see exactly what your draft produces before your instructor does, the most direct way is to run your own Turnitin check on the final file — including both the similarity report and the AI writing report — so there are no surprises at submission time [1].
The best way to understand the AI writing score is to see a real Turnitin AI writing report on your own draft — cover, score, flags, and similarity summary — before you submit. turnitin0 delivers the exact Turnitin AI and similarity reports that match what your university sees, so you can read your score with full context.
※ Turnitin0.com - Actual Turnitin AI Report Cover, Score, Flag And Similarity Summary
FAQ
Does Turnitin detect ChatGPT and other AI tools?
Yes. Turnitin's AI writing detection was designed to identify text produced by large language models such as ChatGPT, GPT-4, and similar AI writing tools, and it reports a percentage of the submission it suspects was AI-generated [1].
What does an asterisk (*%) AI writing score mean?
When the AI writing score is below 20%, Turnitin displays it as *% instead of a single-digit percentage, so the only explicit low numeric outcome a student typically sees is 0% [2].
Is the AI writing score the same as the plagiarism score?
No. The similarity score measures matches against Turnitin's databases, while the AI writing score measures suspected AI-generated text. They are independent indicators that can differ significantly on the same paper [3].
Can a student see the AI writing report for their own paper?
Students should understand how the indicator works before submission, and Turnitin advises instructors to discuss the report with students rather than treat the percentage as a verdict [4].
How can I check my Turnitin AI writing score before submitting?
You can run your own Turnitin check on your final file through a service such as turnitin0, which returns the real AI writing report and similarity report so you can preview your score before your instructor does [1].