OAI-SearchBot
| Vendor | OpenAI |
| Type | AI search index crawler |
| robots.txt token | OAI-SearchBot |
| JavaScript rendering | No — HTTP-only fetcher |
| Honors robots.txt | Yes |
| Vendor docs | platform.openai.com/docs/bots |
User-Agent strings
Mozilla/5.0 AppleWebKit/537.36 (KHTML, like Gecko); compatible; OAI-SearchBot/1.0; +https://openai.com/searchbotPurpose
Builds OpenAI's search index — the corpus that ChatGPT Search uses when a user asks a question that needs current web information. Distinct from GPTBot (training) and ChatGPT-User (live, per-query fetches). Pages that OAI-SearchBot indexes are eligible to appear as sources in ChatGPT's search-grounded answers.
Network identity
- Hostname pattern:
*.openai.com - IP list:
https://openai.com/searchbot.json
In our audit
Blocked at most of the same sites as GPTBot — 62/100 e-commerce sites included OAI-SearchBot in their AI-bot block cluster. The block has a specific cost: those sites are not indexed for ChatGPT Search and therefore cannot appear as cited sources when ChatGPT answers a query that needs fresh web data.
How to allow / block
Allow indexing for ChatGPT Search visibility:
User-agent: OAI-SearchBot
Allow: /Block indexing:
User-agent: OAI-SearchBot
Disallow: /Quirks
- The newest of OpenAI's three bots (announced August 2024).
- Designed to be the AI-search equivalent of Googlebot — sites that
want to appear in ChatGPT Search should allow it.
- Does not render JS; same SSR requirement applies as for GPTBot.