OpenAI
GPTBot
About this crawler
Collects public web content that may be used to train OpenAI’s generative models.
Usually means: OpenAI is gathering content that may improve future models. Disallowing it does not stop OAI-SearchBot from listing you in search.
User agent
GPTBotWe match these as user-agent tokens, not full browser strings. Providers may add surrounding version or product text.
IP verification
Verified when user agent and IP range both match
We first match the crawler’s user-agent token, then compare the request IP with the ranges published by the crawler’s operator.
User agent
The request’s user agent contains GPTBot.
Request IP
Server-side tracking forwards the IP address your server saw.
Range check
The IP is compared with the operator’s published address ranges.
Sample IP ranges
- 132.196.86.0/24
- 172.182.202.0/25
- 172.182.207.0/25
- 20.125.66.80/28
- 20.171.206.0/24
- 20.171.207.0/24
- 4.227.36.0/25
- 52.230.152.0/24
- 74.7.175.128/25
- 74.7.227.0/25
Showing up to 10 of 18 ranges, published 22 September 2026. The live list is fetched from the operator.
Range endpoint
https://openai.com/gptbot.jsonDiscover when bots and AI assistants crawl your website.
Add a small server-side snippet to see which AI assistants, search engines and training crawlers request your pages, including pages that return 404.
- ChatGPT, Claude, Perplexity and Gemini traffic in one view
- Verified against each provider’s published IP ranges
- No visitor cookies and no crawler IP addresses stored