AI content detector

The AI content detector evaluates your text and calculates how much of it has likely been generated by AI. There are two ways you can use this feature.

1. If you have content in a document that you'd like to check, use the free version

You don't need to sign in or create an account to use the free version. Paste or type your text into the text area and select Analyze text.

There's a 60 word minimum and 5,000 word limit to the amount of text you can check at a time. Currently, there's no version available that removes this limit. If you have a long document you'd like to check, we recommend breaking it up and checking 1–2 paragraphs at a time. That way you can narrow in on which parts of your content sound more AI-generated.

2. If you're a developer looking to integrate the AI detector into an app, use the API

Team plan subscribers can check up to 500,000 words per month through the API. To get an API key, select Admin in left menu, then select Account, and then create a service account. The AI content detector is the only API endpoint available on a Team plan.

Enterprise plans have higher limits and full access to the API. If you're interested in detecting AI-generated content at scale, contact our sales team to find out more.

FAQs

Q: I'm logged in to Writer, but I can't find the AI content detector. Where is it?

The AI content detector is only available as a free tool or via the API. You won't find this feature when logged in to Writer.

Q: The detector is scoring something I wrote as AI-generated. Is it accurate?

The AI content detector checks, if given a question or input, whether the text provided is likely to follow the same pattern of words that a large language model would produce. A person's writing can also trigger the detector if they write using very similar word sequences. The detector won't be 100% accurate, but can help give an indication on the likelihood that something was AI-generated.

Q: Can I avoid having my content get flagged as AI-generated?

The distinction between “written by an AI” and “written by a human” isn’t always straightforward. High level AI detectors work by evaluating how “predictable” the text is. In other words, how similar is this text to other texts on the same subject that are already out there. Below are a few ways to avoid your content being flagged as AI-generated:

Making sure that your text is unpredictable — that you’re bringing real value to your readers by saying something that hasn’t already been said before, many times elsewhere
Providing a piece of content that provides unique value to its target audience

Read more about why your content may be flagged as AI-generated.