How to Get Your Website Into ChatGPT Search: OAI-SearchBot, Bing and Mentions
To get your website indexed by ChatGPT, you do not submit it anywhere: you let OpenAI's OAI-SearchBot crawl it, make sure your host or CDN does not block OpenAI's IP addresses, and make the site easy to find through the search providers ChatGPT works with. This guide walks through each step with OpenAI's own documentation, explains what is known (and not known) about Bing, and shows how to check that it worked.

You cannot "register" a website with ChatGPT
There is no ChatGPT equivalent of a submission form or a Search Console where you add your URL. OpenAI's help article on searching the web with ChatGPT describes eligibility in two conditions: allow OAI-SearchBot to crawl the site, and confirm that your website host or content delivery network allows traffic from OpenAI's published search bot IP addresses. The same article says ChatGPT ranks results using multiple factors and that "placement is not guaranteed".
So "getting indexed by ChatGPT" really means three things:
- OpenAI's search crawler can reach and read your pages.
- Your pages can be found through the search providers ChatGPT queries.
- Your site is relevant and trusted enough to be picked as a source for a given question.
This article is about the first two, the technical side. The third, earning citations, is covered in our guide on how to get cited by ChatGPT.
OAI-SearchBot, GPTBot and ChatGPT-User: three bots, three jobs
OpenAI documents its crawlers on one page, Overview of OpenAI crawlers. Each one has its own robots.txt user agent, so you can allow one and block another.
| User agent | What it does | Affects ChatGPT search? | Follows robots.txt? |
|---|---|---|---|
OAI-SearchBot | Surfaces websites in ChatGPT's search features | Yes. OpenAI says sites opted out of it "will not be shown in ChatGPT search answers", though they can still appear as navigational links | Yes |
GPTBot | Crawls content that may be used to train OpenAI's foundation models | No. Blocking it is a training opt-out, not a search opt-out | Yes |
ChatGPT-User | Fetches a page when a user's action in ChatGPT or a custom GPT requires it; not an automatic crawler | Indirectly: it is how ChatGPT opens a page during a conversation | OpenAI says robots.txt rules "may not apply" because a user started the request |
The common mistake is blocking "all AI bots" with one rule copied from a forum. That removes you from ChatGPT search along with training. Decide separately: do you want to appear in ChatGPT answers (OAI-SearchBot), and do you want your content used for training (GPTBot)?
The robots.txt you probably want
For most small sites the sensible default is: allow search, decide about training. A robots.txt that allows ChatGPT search and opts out of training looks like this:
User-agent: OAI-SearchBotAllow: /User-agent: GPTBotDisallow: /
If you are happy for your content to be used for training too, change the last line to Allow: / or simply leave GPTBot out. OpenAI notes that if a site allows both bots, it may use the results of one crawl for both purposes to avoid crawling twice.
Three details people miss:
- A blanket rule counts. If your file has
User-agent: *withDisallow: /and no specific OAI-SearchBot group, the search bot is blocked too. - Changes are not instant. OpenAI says it takes about 24 hours after a robots.txt update for its systems to adjust for search.
- Plugins and hosts edit robots.txt. On WordPress, SEO plugins and some managed hosts generate a virtual robots.txt. Open
yoursite.com/robots.txtin a browser and read what is actually served.
Check your firewall and CDN, not just robots.txt
A perfect robots.txt does nothing if your CDN answers OpenAI's crawler with a challenge page or a 403. Bot protection on Cloudflare, managed WordPress hosts and security plugins can block unfamiliar crawlers by default. This is exactly why OpenAI's eligibility rule mentions hosts and CDNs.
OpenAI publishes the IP ranges for each bot as JSON files, linked from its crawler page: searchbot.json for OAI-SearchBot, gptbot.json for GPTBot and chatgpt-user.json for ChatGPT-User. If you run a firewall, allow OAI-SearchBot by user agent and confirm the request comes from those ranges. User agents are easy to fake; IP ranges are not.
Does ChatGPT use Google or Bing?
This is the most searched question around the topic, and the honest answer is that OpenAI does not publish the full picture. What the primary sources say:
- OpenAI's 2024 launch post, Introducing ChatGPT search, says ChatGPT search "leverages third-party search providers, as well as content provided directly by our partners".
- OpenAI's current help article says ChatGPT search "sometimes partners with other search providers", rewrites your question into targeted queries for them, and points readers to the Microsoft privacy statement (and Shopify's) for how those providers process queries. It does not name a single index.
- Microsoft announced in May 2023 that "Bing is coming to ChatGPT as the default search experience", and ChatGPT's first browsing feature that year worked only via Bing. That is historical context, not a description of how ChatGPT search works today.
The practical reading: Microsoft is one of the search providers ChatGPT works with, and OpenAI also runs its own crawler. How much each contributes is not disclosed. Saying "ChatGPT uses the Bing index" as a flat fact goes beyond the sources, but being properly indexed in Bing is cheap insurance, and it helps with Bing and Copilot on their own.
Get your site into Bing properly
Many small sites have spent years in Google Search Console and never opened Bing Webmaster Tools. It takes a few minutes:
- Add the site. Bing's add and verify guide offers an import from Google Search Console: Bing reads your verified properties and their sitemaps, and imported sites are verified automatically. You can also add a site manually and verify it.
- Check the sitemap arrived. If you imported from Search Console, Bing also picks up new sitemaps from there during its periodic checks. Otherwise, submit your sitemap URL in the Sitemaps section.
- Wait for data. Bing says it usually takes 48 hours to generate analytics for a newly added site.
- Turn on IndexNow. IndexNow lets you notify participating search engines the moment a URL is added or changed, and submitted URLs are shared with all other participating engines. You host a key file at your site root and ping an endpoint, or let a plugin or your CMS do it.
- Fix what Bing reports. Crawl errors, blocked resources and pages excluded from the index show up in Bing's reports just as they do in Google's.
A quick sanity test: search Bing for site:yourdomain.com. If your key pages are missing there, fix that before you worry about ChatGPT.
Make your pages easy to read for a crawler
Once the bot is allowed in, it still has to understand what it finds. The basics are the same as for any search engine, and they matter more for crawlers that may not run your JavaScript the way a browser does:
- Main content in the HTML. If the article text appears only after client-side JavaScript runs, open "view source" and check that the words are really there.
- One clear topic per page, with a descriptive title, a direct answer near the top and headings that match real questions.
- Fast, stable responses. Timeouts and 5xx errors waste a crawl. Our post on Core Web Vitals for small sites covers the fixes that matter most.
- Dates and authorship visible on the page, so a reader (human or not) can judge how current the information is.
- llms.txt is optional. It can help some tools navigate your site, but OpenAI's crawler documentation does not mention it. See our llms.txt guide for when it is worth the effort.
Mentions on other sites are part of being found
Being crawlable makes you eligible. It does not make you the answer. When ChatGPT searches for "best tool for X" or "how do I fix Y", it picks from pages its providers return and from sources it considers reliable. A small site that nobody else mentions is hard to surface for anything beyond its own brand name.
That is why mentions matter: reviews, comparisons, interviews, partner articles and directory listings that name your site in context. We explain the difference in brand mentions vs backlinks, and the step-by-step version is in the guide to getting mentioned by AI assistants. If you want partner articles about your site on similar sites without cold outreach, that is what Satelo's exchange dashboard does: articles are labeled and links use rel="sponsored", so this is about being discussed, not about passing ranking signals.
How to check whether ChatGPT can see your site
There is no ChatGPT index report, so you combine three signals:
- Server logs. Search your access logs for
OAI-SearchBot. Visits from IPs listed in searchbot.json mean the crawler is reaching you. Look at the status codes too: a stream of 403 or 503 responses means something is blocking it. On shared hosting, raw logs are usually under "Logs" or "Raw access" in the control panel. - Manual queries. In ChatGPT, turn on search and ask questions your page answers well, including one with your brand name. Look at the cited sources. Try a few phrasings; results change from query to query.
- Referral traffic. In your analytics, filter referrals from chatgpt.com. Visits there mean people are clicking through from answers.
If OAI-SearchBot appears in your logs with 200 responses but you never show up in answers, the problem is no longer indexing. It is relevance and reputation, which brings you back to content and mentions.
Perplexity and others run their own bots
ChatGPT is not the only AI search engine, and its rules do not carry over. Perplexity documents its own crawlers: PerplexityBot, which surfaces and links websites in Perplexity's search results and which Perplexity says is not used to crawl content for foundation models, and Perplexity-User, which fetches pages for user requests and generally ignores robots.txt. Perplexity also says robots.txt changes can take up to 24 hours. If you want to appear there, allow PerplexityBot the same way you allowed OAI-SearchBot.
Checklist: get your website indexed by ChatGPT
| Step | How to confirm it is done |
|---|---|
| OAI-SearchBot allowed in robots.txt | Live robots.txt has no rule blocking it, directly or via User-agent: * |
| GPTBot decision made | Allowed or disallowed on purpose, not by accident |
| CDN and firewall allow OpenAI's search bot | Log entries from searchbot.json IPs return 200 |
| Site added to Bing Webmaster Tools | Verified, sitemap processed, key pages found with site: |
| IndexNow running | Key file reachable at your root, pings sent on publish |
| Content readable in HTML | Main text visible in "view source" |
| Mentions on other sites | Your brand appears in context on relevant, independent pages |
What to do next
Today: open your live robots.txt, check it against the table above, search your logs for OAI-SearchBot and add the site to Bing Webmaster Tools. Those three steps cover most of the technical reasons a site never appears in ChatGPT search.
Then check your site's basics in one free run: whether it is crawlable, how fast it responds, and whether it is known to Common Crawl and the Tranco list, with the evidence behind each number. It takes about 30 seconds and shows what a crawler sees before you start working on mentions.
FAQ
How do I list or register my website on ChatGPT?
You cannot submit a site to ChatGPT. OpenAI says a site becomes eligible for ChatGPT search when OAI-SearchBot is allowed to crawl it and the host or CDN allows OpenAI's published search bot IP addresses. Placement in answers is not guaranteed.
Does ChatGPT use Google or Bing?
OpenAI says ChatGPT search uses third-party search providers and content from partners, and its help article points to Microsoft's privacy statement for how providers process queries. It does not say that ChatGPT relies on one index, so being well indexed in Bing helps but is not the whole story.
Should I block GPTBot?
Only if you do not want your content used for training. GPTBot is OpenAI's training crawler; blocking it does not remove you from ChatGPT search. OAI-SearchBot is the bot that controls search visibility.
How long does it take to appear in ChatGPT search?
OpenAI says its systems adjust to robots.txt changes in about 24 hours, but there is no published timeline for when a page is crawled or cited. Check your logs for OAI-SearchBot visits to see when crawling starts.
How can I tell if ChatGPT has crawled my site?
Search your server access logs for the OAI-SearchBot user agent and confirm the IP is in OpenAI's searchbot.json list. Then test questions in ChatGPT with search on and look at the cited sources.
Do I need llms.txt to appear in ChatGPT?
No. OpenAI's crawler documentation does not mention llms.txt. It can help some tools navigate a site, but allowing OAI-SearchBot and having readable pages matter far more.
Does blocking OAI-SearchBot remove my site completely?
OpenAI says sites opted out of OAI-SearchBot will not be shown in ChatGPT search answers, but they can still appear as navigational links.
Read next




Check your site's Quality Score
Free, no sign-up: a 0–100 score from mostly open data. Then add your site to Satelo and get partner articles from sites of your level.
Enter a domain like example.com
Add my site