Direct answer
Direct Answer — Turnitin's ChatGPT (AI writing) detection report shows an overall percentage of text likely AI-generated (0%, *% for 1–19%, or 20–100%) along with sentence-level highlights when submitted through an instructor's assignment. However, there is no standalone "free preview" tool for students to check AI scores on their own before submission. What you can access depends entirely on your instructor's settings—some enable AI report visibility, and some institutions offer Turnitin Draft Coach for previewing similarity and AI scores in Google Docs or Word before the final hand-in [1][4].
What Can You See in a Turnitin ChatGPT Detection Report Preview?
When an instructor grants access to the AI writing report, students can view a detailed breakdown that goes far beyond a single number. The report displays an overall percentage indicating how much of the document's qualifying prose text may have been generated by an AI tool such as ChatGPT, GPT-4, or other large language models [1]. Scores below 20% display as an asterisk (*%) rather than a single-digit numeral—a design choice that signals lower reliability in that range—while scores of 0% and 20–100% appear numerically.
The report also provides a sentence-level highlight view with two detection categories: "AI-generated only" (highlighted in cyan) and "AI-generated text that was AI-paraphrased" (highlighted in purple) [2]. This lets you see exactly which passages were flagged and why. An interactive submission breakdown bar gives a visual overview across all pages of the document, allowing you to click on highlighted sections to jump straight to the flagged text. Additionally, instructors can download the report as a PDF, and if visibility is enabled, students can review these same highlights before the due date [1][2].
Importantly, the AI writing detection percentage is independent from the Similarity score. A document can have a high similarity match but zero AI detection, or vice versa, because the two tools analyze entirely different characteristics [2].
How Reliable Is Turnitin's AI Detection for ChatGPT-Generated Text?
Turnitin's AI detection model is trained on a large corpus of both authentic academic writing and AI-generated text across multiple subject areas and geographies, including efforts to minimize bias against second-language learners and underrepresented disciplines [3]. The company reports that for documents with 20% or more AI-detected text, the false positive rate is kept below 1% [1]. However, the model's accuracy decreases in the 1–19% range, which is why scores there are displayed as *% rather than a precise number—to prevent misinterpretation of potentially unreliable data [3].
The model is specifically trained to detect outputs from ChatGPT (GPT-3.5), GPT-4, GPT-4o, Gemini, Claude Sonnet, LLaMA, and other major large language models, and Turnitin continuously updates its detection capabilities as new models emerge [1]. Despite its strengths, Turnitin explicitly states that the AI writing indicator should not be used as the sole basis for academic misconduct decisions. The tool is designed to provide data points that educators evaluate alongside their own professional judgment and institutional policies [1][3].
False positives—incorrectly flagging human-written text as AI-generated—are possible, especially for highly structured or formulaic academic prose. Turnitin advises that any text flagged as AI-generated should be reviewed in context, considering the student's writing history and the assignment's requirements [1].
Can You Check Your Own Work with Turnitin AI Detection Before Submitting to Your Instructor?
Under standard institutional setups, students cannot run their own AI detection check within Turnitin without first uploading to an active assignment created by their instructor. There is no public, self-service portal where a student can paste text and receive an AI score [4]. However, there are two main pathways to gain a preview before the final submission.
Option 1: Turnitin Draft Coach. If your institution has enabled Turnitin Draft Coach, you can check similarity and citation information directly within Google Docs or Microsoft Word as you write. Some institutions also make AI writing insights available through Draft Coach, allowing you to see flags before you submit to the assignment [4]. Check with your school's library or IT department to see if this tool is available.
Option 2: Assignment resubmissions. If your instructor allows resubmissions, you can upload a draft to the assignment, view the generated report, then revise and re-upload. In classic assignments, the first three attempts generate immediate reports; after that, a 24-hour waiting period may apply. In new standard assignments, you can resubmit up to three times within a 24-hour period [4]. This workflow lets you see your AI and similarity scores iteratively before locking in your final version.
If neither option is available, the only way to preview your submission is to ask your instructor to create a separate practice assignment for self-checking purposes [4].
For students who want a reliable way to preview their Turnitin AI detection and similarity scores before submitting to an instructor, Turnitin0 provides a practical self-check solution. Upload your.docx,.pdf, or.txt file and receive a report that matches what your professor sees—including the AI writing percentage, sentence-level highlights, and similarity matches—all without needing Draft Coach or a practice assignment. Results are delivered in minutes, and your paper is never stored or shared with any third party.
※ Turnitin0.com - Actual Turnitin AI Report Cover, Score, Flag And Similarity Summary
FAQ
1. Can I check my Turnitin AI score for free?
There is no official free self-check tool from Turnitin for students. However, if your institution enables Turnitin Draft Coach or you use an assignment's resubmission feature, you can preview your AI detection score at no additional cost [4]. Some third-party services like Turnitin0 also offer a paid self-check option that generates an identical report to your instructor's view.
2. What does the *% mean in the AI writing report?
The asterisk score (*%) indicates that Turnitin's AI detection found between 1% and 19% of the text potentially AI-generated. This range is displayed as an asterisk rather than a specific number because false positive rates are higher below the 20% threshold [1][3].
3. Does Turnitin's ChatGPT detector flag all AI writing the same way?
No. Turnitin distinguishes between two categories: "AI-generated only" (text directly from tools like ChatGPT) and "AI-generated text that was AI-paraphrased" (text modified by paraphrasing tools like Quillbot). These appear as cyan and purple highlights respectively in the report [2].
4. Can I see which sentences were flagged before my instructor does?
Only if your instructor has enabled student visibility for the AI writing report. When enabled, students can see the same sentence-level highlights and percentage that the instructor sees [1]. Without this setting, the report remains visible only to the instructor.
5. Is Turnitin's AI detection accurate for non-native English writing?
Turnitin accounts for second-language learners in its model training to reduce bias, but the detector is optimized for long-form English academic prose. Some research suggests that highly formulaic or simplified writing may trigger false positives. Turnitin explicitly advises using the report as a discussion tool rather than a definitive judgment [1][3].