ChatGPT detector that also checks Claude, Gemini and DeepSeek
IsThisAiWritten is a free ChatGPT detector that also checks text from Claude, Gemini and DeepSeek. Paste your text below and press Check for AI: the check runs in your browser, with no account, no word limit and nothing uploaded.
It works on text from any chatbot
The detector looks at how a text reads, not at which chatbot wrote it. Paste text from ChatGPT, Claude, Gemini, DeepSeek, Grok or any other model and the check runs the same way, with the same outcomes and the same report.
It never says which model wrote a text. The outcome tells you whether the writing reads as AI-written, and nothing about which tool produced it. That suits the usual situation, where you don't know where a text came from, such as a forwarded email or a freelance draft. If you need to know which chatbot was involved, the writer's drafts and their own account of how they worked are the place to look.
Newer model releases
New models arrive often. The home page shows the newest release in each family automatically: today that includes GPT-6.1 Sol, Claude Sonnet 5.5, Gemini 4 Argon and DeepSeek V4.1 Flash. Each release writes a little differently, so results can shift from one version to the next, in either direction. Our post on whether detectors can catch the newest models explains why.
How to check text and weigh the result
- Paste at least 150 words into the box above, or open a .txt or .docx file. Under 50 words there is no result.
- Press Check for AI. Invisible or look-alike characters are set aside before scoring and listed for you.
- Read the outcome (Human, Mixed, AI or Inconclusive) with its confidence word, then the AI-likeness line, which compares the text with the human writing we tested.
- Use the sentence map to find the passages that score highest within the text, and read them in context.
If the outcome is Human but AI-likeness is high, treat the question as still open. Better evidence usually lives outside the text: drafts and version history show how a piece came together, and a short conversation shows whether the writer can explain and extend what they wrote. Surface habits such as em dashes prove nothing on their own; see why em dashes don't prove AI.
What a result from this ChatGPT detector means
Like any detector, it can be wrong in both directions, whichever chatbot wrote the text, so treat a result as a reason to look closer, not as proof. Never use any result as the only evidence against someone. What we offer is a free, private check with no word limit, and our methodology page explains how it is tested.
Questions
Is this detector made by OpenAI, Anthropic, Google or DeepSeek?
No. IsThisAiWritten is independent and has no affiliation with any of them.
Does it check text in Chinese, Spanish or other languages?
No. The detector is calibrated on English only, so other languages return Inconclusive rather than a guess. Text translated into English can be checked.
Can it detect text a person wrote and ChatGPT edited?
Not as a category of its own. A detector built to separate human text from AI text has no label for a blend, so edited text can come back as any of the four outcomes, and none of them says how much the AI contributed.
Is this AI written? Check it now.
Free, no sign-up, no word limit. The check runs in your browser, so your text never leaves your device.
Check text free