AI Detector user guide

The AI detection feature helps you use AI responsibly. It is your authenticity safeguard—it identifies text written and/or modified by major AI models, like ChatGPT, Claude, and Gemini, helping you ensure originality and meet academic, editorial, or institutional standards. Whether you’re reviewing your own or someone else’s work, this feature adds clarity and confidence to the pre-submission and review process.

Depending on your plan, you can access the AI detection feature in different ways. Each entry point offers the same value: Grammarly’s AI detection shows the percentage of text that appears AI-generated and offers guidance on interpreting the results responsibly.

Note: The percentage result should not be used as an objective source of truth, as AI detection of any kind can be prone to errors.

Access AI detection using Grammarly for Windows and Mac

Individual Pro users

To access the plagiarism checker and AI detection feature, click the Grammarly tab on the right edge of your screen and select the Plagiarism agent in the side panel.

Members of team plans

Members of team plans or organizational subscriptions can access the AI detection feature using Grammarly for Windows and Mac only in Microsoft Word. To review your document for AI-generated text, open Grammarly in Word and click the quotation marks icon.

Access AI detection in Google Docs using the Grammarly browser extension

To review your document for AI text in Google Docs, click the Grammarly icon in the lower-right corner and select Detect AI and plagiarism in your text.

Detect AI and plagiarism in your text.png

Access AI detection in Grammarly’s AI writing surface

The AI detection feature is also available as the AI Detector agent in docs, Grammarly’s AI writing surface.

Note: For Grammarly Enterprise, Grammarly Business, and Grammarly for Education users, the AI Detector agent is available in docs if both the agent and the AI writing surface are enabled by the organization’s admin. Admins can adjust agent access for team members on the Agent access page.

To open the AI Detector agent, hover over the agents panel on the right edge of your screen and select AI Detector.

Grammarly will then display the likelihood of parts of your text being flagged as AI-generated, with separate sections marked using the bars on the left side panel:

If you use generative AI when working on your text, AI Detector offers you an option to generate a citation. Click Generate citation to open the citation generator, which will help you create well-formatted APA, MLA, and Chicago-style citations.

As you move through the specific passages that were flagged by the AI Detector agent as possibly AI, you can click Mark as added if you choose to add citations. The count on the right-hand side will update in real time, helping you track where you’ve made modifications to your text and/or added a citation.

Note: If you are working with other Grammarly agents to rewrite your text, it is likely that the AI Detection agent will flag your content as AI-generated. That is because rewrites from the Grammarly, Paraphraser, and Humanizer agents come from our LLM. If you want to avoid this, we suggest you use the agents to receive feedback that you incorporate manually instead of using them to rewrite text directly.

FAQs

What does the % score mean?

The score represents the percentage of scanned text that is likely AI-generated, based on our extensive machine learning training model. Our model is trained on hundreds of thousands of human and AI-generated texts. Grammarly breaks your text into smaller sections, checking each section against the model for patterns that are typically present in AI text (language patterns, syntax, and complexity). Grammarly then returns a percentage score, indicating how much of the provided text appears AI-generated. It assesses whether text resembles AI-generated writing; like other AI text checkers, it cannot provide a definitive conclusion.

How accurate is the score?

The AI detection score is an averaged estimate of the amount of AI-generated text that is likely contained in a given document or piece of writing. Rigorous testing was done to ensure that the model is directionally accurate at identifying AI-generated text. Shorter passages are slightly harder to measure for an accurate score than longer passages.

Like all AI detectors, the AI detection feature is not 100% accurate and should not be used as a definitive assessment of whether AI-generated text is present. The score itself should be viewed as an average estimate rather than a definitive percentage assessment.

I have not used AI-generated text in my document, but Grammarly is still saying that I have AI text in my writing. Why?

The AI detection model is optimized to minimize false positives, as we believe that incorrectly identifying human-generated text as AI-generated is more detrimental than failing to accurately catch AI-generated text. That said, there may be instances in which the human-generated language patterns, syntax, and complexity mirror those of AI-generated text more closely. We expect these instances to be very rare, but as with any AI detector, the results should always be viewed within a broader context and not taken as an objective source of truth.

If my professor uses a different AI detector to scan student writing, will the percentage score be the same?

The AI detection feature uses a proprietary in-house model, so the scores may differ from those of other solutions like Turnitin, GPTZero, Copyleaks, and others. You should not view any scores from Grammarly’s AI detection as a clear indication that your professor will see the same score when they run it through their preferred AI detector. However, our model is trained similarly to some of the other most accurate and used detectors, so we believe the scores should be directionally aligned.

How does AI detection distinguish between AI-generated content, content that is modified by generative AI, and human-written text that is edited with other agents?

Because our AI detection model has been trained solely on human-generated and AI-generated text from LLM providers, traditional, non-generative corrections in the form of Grammarly red and blue underlines should typically not impact the percentage score. However, using more sophisticated Grammarly writing agents in docs, including the Grammarly agent, or using the Grammarly generative AI assistant in Google Docs and Word to meaningfully rewrite sentences and full paragraphs will raise the percentage score. The model is also more likely to assign a higher percentage of AI-generated text in a given passage if it’s been generated by an outside LLM provider (such as ChatGPT or Gemini). That said, paraphrasing edits with Grammarly and other providers are performed by LLMs, so those actions should still lead to some percentage of AI-generated text being triggered in the AI detection.

How can I reduce my document’s AI score?

Currently, we do not offer a dedicated feature to reduce the AI score in user content. Grammarly users who use the AI detection feature should view the results as a data point in deciding how they may want to edit, rephrase, and attribute sources properly to reduce the risk of inappropriate AI use.