WebmasterID logoWebmasterID
AI crawlers

ChatGPT-User — OpenAI real-time fetcher

ChatGPT-User is the token OpenAI uses for real-time fetches made when a person in ChatGPT browses or asks it to read a URL. It is distinct from GPTBot, which crawls for model training, and OpenAI documents both. It honours robots.txt and identifies itself with the ChatGPT-User token plus a self-identifying URL.

Verified against primary sources

What this means

ChatGPT-User represents a real-time fetch OpenAI makes when a ChatGPT user browses the web or asks ChatGPT to read a specific URL. Unlike GPTBot, it is not a background crawl for model training — it is triggered live by a person's request inside ChatGPT.

Because the two have different purposes, OpenAI documents them as separate tokens. Allowing or blocking GPTBot does not change ChatGPT-User behaviour, and vice versa. Decide each policy according to what you actually want to permit.

How ChatGPT-User identifies itself

ChatGPT-User uses the robots.txt user-agent token ChatGPT-User. Its full user-agent string contains that token together with a self-identifying URL pointing at OpenAI's bot documentation. Match on the stable token rather than a full version string, because version components change over time.

As with any crawler, the user agent is a claim and can be copied. For requests where authenticity matters, verify the source against OpenAI's published guidance rather than trusting the user agent alone.

robots.txt considerations

ChatGPT-User honours robots.txt. To disallow it site-wide, target the token specifically:

User-agent: ChatGPT-User Disallow: /

This affects only the real-time fetcher and does not block GPTBot or OAI-SearchBot. robots.txt is a request honoured by compliant clients, not an access-control mechanism.

How it appears in analytics and logs

A request carrying the ChatGPT-User token means OpenAI fetched the URL in real time because a ChatGPT user asked for it — a user-triggered bot fetch, not a background training crawl and not a human page view. Treat it as automated traffic, not audience.

Diagnostic use case

Confirm whether a page was fetched on a ChatGPT user's behalf, and set robots.txt policy for the real-time fetcher separately from the GPTBot training crawler.

What WebmasterID can help detect

WebmasterID classifies ChatGPT-User requests server-side as an AI fetcher and surfaces them on the bot-intelligence and AI-visibility surfaces, so you can see real-time ChatGPT fetches separately from GPTBot training crawls and from human visits.

Common mistakes

Privacy and accuracy notes

Detection uses only the request user-agent. The fetch is triggered by a person, but no visitor identity is exposed to your server beyond a bot request. WebmasterID records it as a bot event, separate from human analytics, and never builds a visitor profile from it.

Frequently asked questions

Is ChatGPT-User the same as GPTBot?
No. GPTBot is a background crawler used to help train OpenAI's models. ChatGPT-User is a real-time fetch made when a ChatGPT user browses or asks it to read a URL. They use different robots.txt tokens and are controlled independently.

Related pages

Sources and verification notes

Last reviewed 2026-06-24. Facts are checked against primary/official sources where available; uncertain specifics are marked “Data not yet verified” rather than guessed.