Assistant fetchers
ChatGPT-User is a bot operated by OpenAI that fetches single pages on demand for an AI assistant. ChatGPT-User is OpenAI's web crawler that visits websites when ChatGPT users request information. It honours robots.txt, and a request claiming to be ChatGPT-User can be checked against a published IP range file.
Request says
"ChatGPT-User/1.0"
Checked against
OpenAI's published IP ranges
Live check
Paste the address from the log line. It is checked against what the operator publishes.
Reference
| User agent | Mozilla/5.0 AppleWebKit/537.36 (KHTML, like Gecko); compatible; ChatGPT-User/1.0; +https://openai.com/bot |
|---|---|
| Operator | OpenAI |
| Purpose | Assistant fetchers |
| Respects robots.txt | Honours |
| Verification method | IP rangesPublished IP range file Fetches are triggered by a person asking ChatGPT about a URL, so traffic is spiky rather than a steady crawl. The egress ranges are published as JSON. |
| Crawl pattern | Only when prompted by a user. |
| Robots.txt tokens | ChatGPT-User |
| Operator documentation | https://platform.openai.com/docs/bots |
ChatGPT-User is OpenAI's web crawler that visits websites when ChatGPT users request information. This enables ChatGPT to include links in its responses.
robots.txt
Paste one at the root of your domain. Rules are per token.
User-agent: ChatGPT-User
Disallow: /User-agent: ChatGPT-User
Allow: /User-agent: ChatGPT-User
Allow: /
Disallow: /account/
Disallow: /checkout/
Disallow: /searchEdit the Disallow paths to match your own URLs.
Blocking ChatGPT-User means a person who asks about one of your URLs gets an error instead of a summary. That is a support and reputation trade-off, not a training one.
Build a file for every crawlerFAQ
ChatGPT-User is a bot operated by OpenAI that fetches single pages on demand for an AI assistant. ChatGPT-User is OpenAI's web crawler that visits websites when ChatGPT users request information. It honours robots.txt, and a request claiming to be ChatGPT-User can be checked against a published IP range file.
Yes. The operator documents that ChatGPT-User reads robots.txt and follows Disallow rules for its own token. Rules take effect on the next fetch of the file, which can be hours or days after you publish them.
Fetches are triggered by a person asking ChatGPT about a URL, so traffic is spiky rather than a steady crawl. The egress ranges are published as JSON. The reference is published at https://openai.com/chatgpt-user.json.
Add the ChatGPT-User token to robots.txt with Disallow: /. If the crawler ignores robots.txt, match the user agent at your CDN or reverse proxy and return 403 there instead.
Blocking ChatGPT-User means a person who asks about one of your URLs gets an error instead of a summary. That is a support and reputation trade-off, not a training one.