Direct answer
Direct Answer - ** A Turnitin check compares your submitted document against a large database of internet content, academic publications, and previously submitted student papers, then produces a Similarity Report showing which passages match existing sources [1]. Most modern Turnitin checks also run AI-writing detection**, which estimates how much of your text appears to be generated by AI tools [2]. Together, these two signals — similarity and AI score — are indicators your instructor reviews, not automatic verdicts of misconduct [1][3].
What does a Turnitin check actually look for in your document?
A standard Turnitin check runs your text against three source categories: web pages and other internet content, periodicals and scholarly journals, and a repository of papers previously submitted by students [2]. Every passage that overlaps with one of those sources is highlighted and attributed, so the reader can see exactly where a match came from [1]. The result is a similarity percentage plus a source-by-source breakdown, not a single pass/fail grade [1].
Alongside similarity, the check examines whether the writing itself looks machine-generated [2]. This AI-detection layer estimates the proportion of qualifying text likely produced by a large language model, and flags the specific segments it considers suspicious [2]. Both checks run on the same submission, which is why a single report can carry two very different numbers [1][2].
What do the similarity and AI scores in a Turnitin report mean?
The similarity score is a percentage of your document's text that matched existing sources — it is a measure of overlap, not a measure of originality or honesty [1]. A high score can mean heavy copying, but it can equally reflect a topic saturated with standard terminology or a reference list that was not excluded [1]. Turnitin deliberately avoids labelling any score as "good" or "bad," because context changes the meaning entirely [1].
The AI score works differently. It estimates how much of the qualifying text appears AI-generated and highlights the flagged passages so a reader can inspect them directly [3]. When Turnitin's confidence is low, the report displays an asterisk (*%) instead of a precise percentage — this happens when the AI signal falls below its 20% confidence threshold [3]. A score shown as *% therefore indicates an uncertain, low-confidence detection rather than a confirmed result [3].
Neither number is proof on its own. Turnitin states that both the similarity percentage and the AI percentage are indicators intended to prompt review and conversation, not automated accusations [3]. The practical takeaway is that you should read the flagged segments, not just the headline figure, because the context of each match is what actually determines whether there is a problem [1][3].
How can you see your own Turnitin AI and similarity results before you submit?
This is the gap most students hit. Turnitin's own help documentation explains that students generally cannot run an official institutional check on a draft before submitting — the similarity and AI reports are generated inside the instructor's assignment and are visible to the instructor, not to the student beforehand [4]. In other words, the official check happens at submission, which is often too late to revise [4].
Because of that limitation, many students turn to independent pre-submission services that generate Turnitin AI and similarity reports on their own draft [4]. The value is timing: you see the flagged passages while you can still rewrite them, instead of discovering a problem after the deadline has passed [4]. This is especially relevant if any part of your draft was drafted or polished with an AI assistant, since AI-detection flags are exactly what a preview surfaces [2][4].
A pre-submission preview does not replace your institution's official submission, and it does not change what your instructor will see [4]. What it does is give you the same two signals — similarity matches and AI flags — early enough to act on them [1][4]. For a student who is unsure whether their draft will raise questions, that early visibility is the whole point [4].
If you would rather not wait until submission to find out what your Turnitin report says, turnitin0 lets you preview the real thing on your own draft first — so the numbers you see are the numbers your instructor will see.
※ Turnitin0.com - Actual [Turnitin AI Report](https://www.turnitin0.com/guides/us/turnitin-paid-account) Cover, Score, Flag And Similarity Summary
Importantly, the check does not judge intent. A correctly quoted line, a common phrase, or a properly cited definition can all raise your similarity score without any wrongdoing [1]. That is why Turnitin's own guidance frames the Similarity Report as a tool for interpretation rather than a plagiarism detector [1]. Depending on how the assignment is configured, your instructor may also exclude elements like the bibliography or quoted material from the calculation [2].
FAQ
Does a Turnitin check tell my instructor I cheated?
No — it produces similarity and AI percentages that an instructor interprets in context, not a misconduct ruling [1][3]. A high score simply means text matched a source or read as machine-generated, and it can have innocent explanations [1].
What is the difference between the similarity score and the AI score?
The similarity score measures how much of your text overlaps with existing sources, while the AI score estimates how much of it appears AI-generated [1][3]. They are calculated independently and appear as two separate numbers on the same report [2].
Why does my AI score show an asterisk instead of a number?
Turnitin displays *% when its AI detection confidence is low — specifically when the signal falls below its 20% confidence threshold [3]. It signals an uncertain result rather than a confirmed one [3].
Can I check my own draft with Turnitin before I submit it?
Turnitin's own documentation notes that students generally cannot run an official institutional check beforehand, because reports are generated inside the instructor's assignment [4]. Independent pre-submission services exist precisely to give students that early view [4].
Does a pre-submission check affect my official Turnitin submission?
No — a preview on your own draft does not alter what your instructor sees when you submit through your institution [4]. It only gives you the similarity and AI signals earlier so you can revise flagged passages [1][4].