ChatGPT-User
Fetches a URL right now because a ChatGPT user just asked a question that requires it — clicking a link, asking ChatGPT to read a specific page, or letting the 'browse' feature follow a search result. Every request maps to an active human-driven query.
Specs
| Vendor | OpenAI |
| Category | FETCH |
| robots.txt token | ChatGPT-User |
| Renders JavaScript | partial JS |
| Honors robots.txt | yes |
| Reverse-DNS pattern | *.openai.com |
| IP-range source | https://openai.com/chatgpt-user.json |
User-Agent string
Mozilla/5.0 AppleWebKit/537.36 (KHTML, like Gecko); compatible; ChatGPT-User/1.0; +https://openai.com/botConsiderations
- This is the bot that directly affects revenue: blocking it removes the retailer from ChatGPT shopping answers in real time.
- Uses a headless browser tier for some flows (browse / read a page); other flows are simple HTTP fetches.
- OpenAI's docs explicitly state ChatGPT-User fetches are 'made for users' and not used for training. Sites that block it on anti-training grounds are misreading the bot.
robots.txt recipe
User-agent: ChatGPT-User
Allow: /Sources: OpenAI · Crawler overview