OAI-SearchBot: the crawler that makes you visible in ChatGPT
OAI-SearchBot is OpenAI's robot building the ChatGPT search index: it's what allows a website to appear, with a clickable link, in ChatGPT answers when it searches the web. OpenAI states it is not used to train models. For GEO visibility, it is the most important OpenAI bot to allow.
Technical profile
- Operator
- OpenAI
- Role
- ChatGPT search index
- robots.txt token
OAI-SearchBot- User-agent (example)
Mozilla/5.0 AppleWebKit/537.36 (KHTML, like Gecko); compatible; OAI-SearchBot/1.0; +https://openai.com/searchbot- Respects robots.txt
- Yes (documented by OpenAI)
- Official IP ranges
- 336 ranges published by the operator across all its bots (snapshot of 2026-07-23, refreshed on every deployment) — official file
- Official documentation
- OpenAI bots documentation
Why OAI-SearchBot matters for GEO
When a ChatGPT user asks a factual or current question, the engine queries its search index and cites sources with links. That index is built by OAI-SearchBot. A site it cannot crawl simply doesn't exist in ChatGPT search — regardless of content quality.
It mirrors the Googlebot ↔ Google Search relationship, transposed to ChatGPT. The difference: OpenAI's ecosystem explicitly separates training (GPTBot) from search (OAI-SearchBot). You can refuse one while cultivating the other — a training opt-out without sacrificing visibility.
OAI-SearchBot in your logs
Its presence in your logs is good news: it means ChatGPT is indexing your content. Watch which pages it visits — those are the ones with a chance of being cited. If it ignores your strategic pages, work on internal linking and citability (self-contained answers, sourced data).
Verifying it's really OAI-SearchBot
Like all OpenAI bots, OAI-SearchBot can be verified through its official IP ranges, published in a dedicated JSON file. An “OAI-SearchBot” whose IP falls outside those ranges is an impostor. Our analyzer runs this check automatically, in your browser.
View OAI-SearchBot's official IP ranges (JSON)Blocking or allowing OAI-SearchBot (robots.txt)
Unless you have a strong reason, let OAI-SearchBot access your site: blocking it removes you from ChatGPT search, one of the fastest-growing discovery channels. If you only want to refuse training, block GPTBot — not OAI-SearchBot.
To block it:
User-agent: OAI-SearchBot Disallow: /
To explicitly allow it:
User-agent: OAI-SearchBot Allow: /
Frequently asked questions
What's the difference between OAI-SearchBot and GPTBot?
GPTBot collects content to train OpenAI's models; OAI-SearchBot builds the search index that surfaces and cites sites in ChatGPT. OpenAI documents that OAI-SearchBot is not used for training. You can block one without the other: two distinct robots.txt tokens.
Should I block OAI-SearchBot?
For almost every site, no: blocking it removes your visibility in ChatGPT search with nothing gained in return. The only reasonable use case is content you don't want appearing in any engine, AI or classic.
How do I know if ChatGPT indexes my site?
Look for OAI-SearchBot in your server logs: its visits mean the ChatGPT search index is discovering your pages. Our log analyzer isolates it automatically and verifies its authenticity by IP, without uploading your files.
Sources
Every fact on this page is backed by a primary source — the operator's official documentation and, where it exists, its published IP list.
See what this bot does on your site
Drop your access.log files into the Crawl Lab log analyzer: it isolates OAI-SearchBot, verifies its authenticity and lists the pages it requests — entirely in your browser, no upload.
Analyze my logs