AI crawlers · OpenAI
Is your site blocking GPTBot?
GPTBot is OpenAI's crawler for model training. Check your site now: we read your robots.txt exactly as GPTBot must, ask your server as each AI crawler, and look at your product pages.
Does blocking it hurt?
Model training Blocking GPTBot only opts your content out of OpenAI's model training. It doesn't remove you from OpenAI's AI answers (that's OAI-SearchBot). Whether to allow training is your choice, not a fault.
In our research, 4% of 2,159 online shops block GPTBot in robots.txt. See the research.
Allow or block it in robots.txt
OpenAI says GPTBot follows robots.txt, so the lines below decide whether it reads your shop.
To let it in (add to your robots.txt, at yourshop.com/robots.txt):
User-agent: GPTBot Allow: /
To keep it out:
User-agent: GPTBot Disallow: /
A more specific group wins: if your robots.txt also has a group for GPTBot, that group decides, not the “User-agent: *” rules. Our check reads it the same way (RFC 9309).
Other places it gets blocked
- Cloudflare. Cloudflare has blocked AI crawlers by default on new sites since July 2025, and it can also write your robots.txt for you. If your site uses Cloudflare, check its AI crawler settings first (MIT Technology Review).
- Security plugins and firewalls. WordPress security plugins and host firewalls sometimes refuse crawlers they don’t recognise, even when robots.txt allows them. Our check asks your server as each crawler and shows any refusal.
Where this comes from
OpenAI’s own documentation: developers.openai.com. OpenAI: disallowing GPTBot signals content should not be used to train models; it does not affect ChatGPT search.
OpenAI’s other crawlers
- OAI-SearchBot (AI search and answers)
- ChatGPT-User (Opens pages when a person asks)
We check 18 AI crawlers in one go. See them all · How the check works