What is OAI-SearchBot?
OAI-SearchBot is OpenAI's web crawler that fetches and indexes pages so they can be cited as sources in ChatGPT search results. It is 1 of three separate OpenAI crawlers, and it is distinct from GPTBot, which collects content for model training. Because the user-agents are listed independently in OpenAI's bot documentation, the rule is simple: “you can allow OAI-SearchBot while disallowing GPTBot for training.” If you want to appear as a source in ChatGPT, this is the bot to let in.
The short definition
When ChatGPT answers a question using live information from the web, it does not browse your site the way a person does. It relies on an index built by a crawler, and the crawler that builds that index is OAI-SearchBot. It requests your pages, reads the HTML your server returns, and stores what it finds so ChatGPT can pull a quote and a link when the topic comes up. No OAI-SearchBot access, no entry in that index, no citation.
How it fits with OpenAI's other crawlers
OpenAI runs three named crawlers, and confusing them is the most common reason a site blocks itself out of ChatGPT search by accident. They do different jobs and are controlled separately in robots.txt:
- OAI-SearchBot fetches and indexes pages so they can be cited as sources in ChatGPT search results. This is the one that earns you a citation.
- ChatGPT-User fetches a single page in real time when a user explicitly asks ChatGPT to open a link. It acts on a direct request, not a crawl.
- GPTBot collects content to train future OpenAI models. It has nothing to do with whether you get cited, and blocking it does not cost you search visibility.
The important consequence is that these are independent. You can welcome the bot that cites you and refuse the bot that trains on you. A blanket rule that blocks every OpenAI agent to "keep the AI out" also blocks the crawler that puts you in the answer, which is usually the opposite of what the owner wanted.
Why it matters for citations
Getting cited in ChatGPT search is a two-part problem, and OAI-SearchBot is the first part. If the crawler cannot reach and read your page, nothing else you do matters, because there is no version of your content in the index to quote. We see this constantly: a site looks fine in a browser, ranks in Google, and is completely absent from ChatGPT because one line in robots.txt turned the search bot away. Letting OAI-SearchBot in does not guarantee a citation, but blocking it guarantees the opposite.
The second part is whether the page is worth quoting once the bot can read it. Serve your answer in real HTML rather than rendering it with JavaScript after load, lead with a specific, self-contained passage, and include a number or a named entity near the top. Our full walkthrough lives in how to get cited by ChatGPT, and the deeper robots.txt fix is in how to fix robots.txt blocking AI crawlers.
How to check and allow it
Open your robots.txt and make the split explicit. Allow the bots that put you in answers, decide for yourself on the training bot:
# Let ChatGPT cite you in search results User-agent: OAI-SearchBot Allow: / User-agent: ChatGPT-User Allow: / # Optional: keep your content out of model training User-agent: GPTBot Disallow: /
Two things to verify after you publish that file. First, confirm there is no broader Disallow earlier in the file that overrides the allow, since the most specific matching group wins and a stray wildcard rule can quietly shadow your intent. Second, watch your server logs. The user-agent token OAI-SearchBot appears in the request logs when the crawler visits, so a log search for that string tells you whether it is actually reaching your pages or being turned away at the door.
You can check all of this by hand, or paste your link into our AEO checker and let Brimm read your site the way the engines do. We report whether OAI-SearchBot can reach you, whether your answer survives without JavaScript, and how quotable your top passage is. When you want the full picture across every AI crawler and the fixes in order, run the audit at Brimm and start with the fix library.
See also
OAI-SearchBot is one crawler in a field of them. Perplexity runs its own, described in what is PerplexityBot, and the same allow-the-right-bot logic applies across all of them. If your current robots.txt is the thing standing between you and a citation, the step-by-step repair is how to fix robots.txt blocking AI crawlers.