ChatGPT-User
Fires when a ChatGPT user asks about a specific page and the assistant goes to read it live. Each hit represents a real person, mid-conversation, being shown your content - the clearest demand signal in this table.
What it feeds
ChatGPT-User fetches a page during a conversation, when a person asks ChatGPT or a custom GPT something that requires reading that page.[1] Every hit is one human request, which makes the token a demand signal rather than a crawl statistic.
How to identify it
Blocking and its consequences
OpenAI documents that this token does not crawl automatically[3] and that robots.txt rules may not apply to a fetch a user started.[4] A disallow is therefore an unreliable block, and blocking it at the edge removes the site from live ChatGPT answers for the people who asked for it by name.
Verification
Operator documentation: https://platform.openai.com/docs/bots
Published IP list: https://openai.com/chatgpt-user.json
A user agent is a claim, not a proof: any client can send any name. Where an operator publishes IP ranges, a reverse lookup is the only way to tell a real visit from an impersonation.
Seen by Baseline
Requests from this bot across sites tracked by Baseline, last 30 days. Aggregated over - sites, never reported per site.
References (4)
Last updated 2026-09-05. Written and maintained by Baseline Labs.
