Direct answer
Direct Answer - ** A Turnitin checker score is the pair of numbers on a Turnitin report: the AI writing percentage (how much of the qualifying prose the detector flags as likely AI-generated) and the similarity index** (how much of the text matches other sources in the repository) [1]. They are two independent measurements, not one combined grade, and neither is a verdict on academic misconduct — Turnitin presents them as signals for a human to interpret [1]. A score of 0% similarity or an asterisk in place of an AI number both mean "below the confidence threshold," not "guaranteed safe" [3].
What Does a Turnitin Checker Score Actually Mean?
A Turnitin checker score is not a single grade. A report contains a similarity score, which is the percentage of a paper's text that matches sources held in the comparison repository, and a separate AI writing percentage, which estimates how much of the qualifying prose reads as machine-generated [1]. The similarity figure deliberately excludes bibliography entries and quoted material by default, so a paper full of correctly attributed quotations can still post a high number [2]. Reading the headline percentage without opening the match breakdown is the single most common way students misread their own report [2].
The similarity score is best understood as a map, not a mark. Each highlighted passage links to the specific source it matched, which lets you confirm whether a hit is a properly cited quotation, a common phrase, or genuinely unattributed text [2]. Turnitin's own guidance tells markers to treat a high similarity score as a prompt to inspect the matches rather than as proof of plagiarism [2]. That distinction matters because a 30% similarity score built from a reference list and cited theory is a very different document from a 30% score built from copied paragraphs [2].
The AI writing percentage works on a different principle entirely. Turnitin's detector segments the submitted text and evaluates sentence-level patterns, then reports the share of qualifying prose it classifies as AI-generated [1]. It is an estimate produced by a statistical model, and Turnitin explicitly frames it as one data point among several rather than a determination [1]. Because the two scores come from unrelated methods, a paper can show 5% similarity and 80% AI, or 45% similarity and 0% AI — the combination tells you which problem to investigate [1].
What Is a Good Turnitin AI Score, and What Do the Percentage Thresholds Mean?
There is no official "pass" number, but the display rules tell you a lot about how to read one. Turnitin shows an asterisk (*%) instead of an exact figure when its AI detection falls below the model's confidence threshold, which is why a report can show a symbol where you expected a percentage [3]. That asterisk is a low-confidence signal, not a clean bill of health and not an accusation — it means the detector did not find enough consistent evidence to commit to a number [3]. A precise percentage, by contrast, means the model did cross its confidence bar for the flagged portion of the text [3].
The detector also depends on how much qualifying prose exists to analyse. Very short submissions — a paragraph or a page — may return no AI score at all, because there is not enough text for sentence-level pattern analysis to be reliable [3]. This is why two students can submit near-identical work and see different results: length, structure, and how much of the document counts as "qualifying prose" all shift the output [3]. Turnitin itself warns that detectors can produce both false positives and false negatives, and instructs instructors to use the score as a starting point for a conversation rather than a conclusion [3].
In practice, most institutions treat a low or asterisked AI score as unremarkable and reserve scrutiny for high percentages on long, heavily flagged passages [3]. If you see a number you did not expect, the productive move is to look at which sentences were flagged and whether they match the passages you drafted with an AI assistant [1]. Rewriting those specific passages in your own voice, and keeping your prompt history, addresses the underlying signal instead of the number itself [1].
Can Students Check Their Turnitin Score Before Submitting?
Sometimes — and it depends entirely on your institution. Turnitin's own student-facing guidance frames the tool as a formative aid for revising drafts, but whether you can run a self-check is controlled by whether your instructor has enabled draft or pre-submission access for that specific assignment [4]. Many universities switch it off for graded submissions precisely so the report cannot be gamed before the deadline [4]. If the draft folder is not there, no amount of navigating the Turnitin interface will produce a score for you [4].
Where institutional access is unavailable, students commonly look for an independent way to preview the same two figures before the graded submission goes in [4]. The practical value is the same either way: you see the AI percentage and the similarity index, read the match breakdown, and fix the specific passages that triggered a flag while you still have time to revise [4]. Reviewing a report before submission turns a score from a verdict into feedback, which is exactly the formative use Turnitin's student guidance describes [4].
A few habits make a self-check genuinely useful rather than just reassuring. Check the report as a whole — AI percentage, similarity index, and the flagged passages together — because a clean similarity score says nothing about AI detection and vice versa [1]. Compare the flagged sentences against your own drafts so you know which parts of your process produced the signal [4]. And re-check after revising, since a second report on the edited document is the only way to confirm the flagged passages actually changed [4].
If you want to see those exact two numbers — the AI writing percentage and the similarity index — on your own draft before the graded submission, turnitin0 gives you the same report format your instructor sees, so you can read the flagged passages and revise with time to spare.
※ Turnitin0.com - Actual Turnitin AI Report Cover, Score, Flag And Similarity Summary
FAQ
Does a 0% Turnitin score mean my work is safe?
No. A 0% similarity score only means no text matched the repository, and an asterisked AI score only means the detector fell below its confidence threshold [3]. Neither figure rules out a concern an instructor might raise on other grounds [1].
Why does my Turnitin report show an asterisk instead of an AI percentage?
Turnitin displays *% when AI detection falls below the model's confidence threshold, so no precise figure is shown [3]. It is a low-confidence signal rather than a pass or a fail [3].
Is a high similarity score the same as plagiarism?
Not automatically. The similarity index counts matched text and excludes bibliography and quoted material by default, so correctly cited sources still raise the number [2]. You have to open the match breakdown to see what actually matched [2].
Can I check my Turnitin score before I submit?
Only if your institution has enabled draft or pre-submission access for that assignment [4]. Where it is not enabled, students often preview the same AI and similarity reports through an independent checker before the graded submission [4].
Do the AI percentage and similarity index measure the same thing?
No. The similarity index measures matched text against the repository, while the AI percentage estimates how much qualifying prose reads as machine-generated [1]. They come from different methods, so one can be high while the other is low [1].