Direct answer
The Turnitin scanner is the underlying engine that powers both the Similarity (plagiarism) check and the AI writing detection report on submitted academic papers. When a document is uploaded, Turnitin's scanner processes it through two independent but complementary systems: the text-matching database comparison for similarity and a machine learning model for AI-generated text detection [1]. These scans run simultaneously, generating two separate reports that help educators assess authenticity and originality. For students, understanding how the Turnitin scanner works is essential for navigating submission requirements and maintaining academic integrity.
What Is a Turnitin Scanner and How Does It Scan Documents?
The Turnitin scanner refers to the automated system that analyzes submitted documents by breaking them into segments and evaluating each segment against Turnitin's detection models [1]. The scanning process begins when a paper is uploaded — the system first divides the document into segments of roughly a few hundred words, approximately five to ten sentences each. These segments are overlapped with one another to ensure every sentence is evaluated in its full context [1].
Once segmented, each text block is run against Turnitin's AI detection model, which assigns a score between 0 and 1 to each sentence. A score of 0 indicates the model found no evidence of AI generation, while a score of 1 means the model identified the sentence as entirely AI-generated [1]. These individual sentence scores are averaged across the document to produce an overall AI writing percentage.
The scanner evaluates word probability sequences — AI-generated text tends to select the next word in a highly consistent and predictable fashion, whereas human writing is naturally more idiosyncratic and variable in word choice [1]. The model was trained on a representative sample of both AI-generated and authentic academic writing across geographies, subject areas, and language backgrounds, including second-language learners and underrepresented groups, to minimize bias [1]. The Similarity Report side of the scanner runs simultaneously, comparing text against a massive database of web pages, academic journals, and previously submitted student papers [2].
What Does a Turnitin Scanner Check For — Similarity or AI Writing?
The Turnitin scanner checks for both similarity and AI writing, but these are two entirely separate detection engines running on the same submission [1]. The Similarity Report identifies text that matches existing sources — web content, academic publications, and student paper repositories — producing a percentage that represents how much of the paper matches other works. The AI writing detection report, by contrast, identifies text likely generated by large language models such as GPT-4, Claude, Gemini, and others [1].
These two indicators are fundamentally different and should never be confused with one another. A high similarity score means the text closely matches existing sources (potential plagiarism), while a high AI score means the text exhibits patterns consistent with AI generation [3]. A paper can have a high similarity score but a low AI score, or vice versa — the engines operate independently.
Turnitin's AI writing detection capabilities have expanded significantly since launch. The first iteration could detect GPT-3, GPT-3.5, and ChatGPT, but the current system has been updated to detect GPT-4, GPT-4o, GPT-5, Gemini variants, Claude Sonnet models, LLaMA, and many other large language models [1]. The scanner also includes AI paraphrasing detection and AI bypasser detection to identify text that was rewritten using AI paraphrasing tools or passed through humanizer services [1].
It is important to note that in standard institutional Turnitin setups, only instructors and administrators can see the AI writing indicator and report — students generally do not have direct access to the AI detection report through their university's LMS [1][2].
How Can Students Use a Turnitin Scanner to Check Their Work Before Submission?
For students who want to check their work using a Turnitin scanner before submitting to an official assignment, the options depend on their institution's configuration. Some universities provide Turnitin Draft Coach, a tool that allows students to run Similarity Reports, citation checks, and grammar checks directly within Google Docs or Microsoft Word before submitting [2]. If Draft Coach is available, students can preview their similarity results without uploading to a formal assignment.
In cases where Draft Coach is not provided, students can check their work by submitting drafts to an assignment that allows resubmissions. However, this approach has limitations — Classic Standard Assignments allow immediate reports for the first three attempts, after which a 24-hour waiting period applies. New Standard Assignments limit students to three resubmissions per calendar day [2].
Given these institutional restrictions, many students turn to third-party Turnitin checking services that deliver both the Similarity Report and the AI writing report without the constraints of assignment-based checking [4]. These services allow students to scan their documents, preview their AI score and similarity percentage, and make revisions before the final submission — all without waiting for instructor permissions or dealing with resubmission limits.
Pre-submission scanning gives students valuable insight into how their work will appear from the instructor's perspective. Understanding your AI score and similarity percentage before the deadline allows you to revise flagged sections, add proper citations, and ensure your writing reflects your own authentic voice [4].
The best way to understand exactly what the Turnitin scanner will detect on your paper is to see the actual report before you submit. Turnitin0.com provides real Turnitin AI and Similarity reports — the same format and scoring system instructors see in their academic dashboard — so you can preview your scores, identify flagged content, and make informed revisions with confidence.
※ Turnitin0.com - Actual Turnitin AI Report Cover, Score, Flag And Similarity Summary
FAQ
Can students run a Turnitin scanner on their own before submitting an assignment?
In standard institutional Turnitin setups, students cannot directly run the AI writing detection scanner on their own papers — the AI detection report is only visible to instructors. However, some institutions provide Draft Coach for pre-submission similarity checking, and third-party services like Turnitin0.com offer both Similarity and AI detection reports for students who want to preview their scores [2].
Does the Turnitin scanner detect ChatGPT, Claude, and Gemini?
Yes. Turnitin's AI detection model has been continuously updated and currently detects text generated by GPT-4, GPT-4o, GPT-5, Claude Sonnet, Gemini (including Pro and Flash variants), LLaMA, and many other large language models [1]. The scanner also identifies AI-paraphrased text and text that has been passed through AI bypasser tools.
Is the Similarity score the same as the AI score?
No. The Similarity score measures how much text in a document matches existing sources — web pages, journals, and other student papers. The AI score measures whether text was generated by an AI writing tool. These are independent indicators produced by separate scanning engines [1][3].
What happens if my Turnitin AI score is below 20%?
In Turnitin's AI writing report, any score below 20% is displayed as *% (the asterisk bucket) rather than as a specific single-digit percentage. The only explicit low numeric outcome shown is 0%. This means scores such as 3%, 12%, or 18% all appear as the asterisk indicator [1].
How long does it take for the Turnitin scanner to process a document?
In most cases, the Turnitin scanner delivers results within 5–10 minutes. The system processes both the Similarity and AI detection scans simultaneously on the same submission. In rare cases, delivery may take up to 30 minutes depending on system load and document complexity [1].