AI search

Can ChatGPT and Google AI read your Facebook page? What AI assistants can and can't see

[Author name], iPlan Digital
Updated · 7 min read

Key takeaways

  • On 27 October 2026, Facebook, Instagram, TikTok, X and LinkedIn all blocked GPTBot, PerplexityBot, ClaudeBot and Google-Extended in robots.txt.
  • OpenAI's search crawler, OAI-SearchBot, was blocked on four of the five; LinkedIn allowed it to parts of the site.
  • Googlebot was allowed on all five, so public social pages can still appear in Google results and its AI features.
  • User-triggered fetchers such as ChatGPT-User may ignore robots.txt, but logins and bot checks still stop them on most platforms.
  • Friends-only posts, private groups, Stories and chats are invisible to every crawler. For AI search, your facts need to live on your own website.

Can AI read Facebook? Mostly, no. When we checked on 27 October 2026, Facebook’s robots.txt blocked the crawlers that ChatGPT, Perplexity and Claude use, and so did Instagram, TikTok, X and LinkedIn for most of them. Google is the exception: Googlebot can crawl many public social pages, so Google’s AI features may still see a public Facebook page. But anything behind a login (friends-only posts, private groups, Stories, chats) is invisible to every AI crawler.

How AI assistants get their information

An AI assistant answering a question about businesses usually searches an index of web pages and summarises what it finds, with links. Each company runs its own crawlers, and each crawler has a job:

Crawler Company What it’s for
Googlebot Google Google Search, including AI Overviews and AI Mode
Google-Extended Google Not a separate crawler but a token: controls whether content is used to train Gemini models and to ground answers in Gemini Apps. It doesn’t affect Google Search
OAI-SearchBot OpenAI ChatGPT search results. Sites that block it don’t appear in ChatGPT search answers
GPTBot OpenAI Collecting content that may be used to train OpenAI models
ChatGPT-User OpenAI Fetching a page when a user asks; OpenAI says robots.txt rules “may not apply”
PerplexityBot Perplexity Perplexity search results (not training)
ClaudeBot Anthropic Collecting content that may be used for training

Websites tell these crawlers what they may fetch in a public file called robots.txt. That makes it possible to check exactly what the big social platforms allow.

Can AI read Facebook and other social media? Platform by platform

We downloaded the live robots.txt files of Facebook, Instagram, TikTok, X and LinkedIn on 27 October 2026. “Blocked” means the crawler is told not to fetch any page (Disallow: /). “*” is the default rule for any crawler not named in the file.

Crawler Facebook Instagram TikTok X LinkedIn
GPTBot Blocked Blocked Blocked Blocked (via *) Blocked
OAI-SearchBot Blocked (via *) Blocked (via *) Blocked Blocked (via *) Allowed, many paths excluded
ChatGPT-User Blocked (via *) Blocked (via *) Blocked Blocked (via *) Blocked
PerplexityBot Blocked Blocked Blocked Blocked (via *) Blocked
ClaudeBot Blocked Blocked Blocked Blocked (via *) Blocked
Google-Extended Blocked Blocked Blocked Blocked Blocked
Googlebot Allowed, ~40 paths excluded Allowed, ~10 paths excluded Allowed to much of the site (via *) Allowed, some paths excluded Allowed, many paths excluded
* (all other bots) Blocked Blocked Partly allowed Blocked Blocked

A few details from the files:

  • Facebook and Instagram use the same structure. Both start with a notice that automated data collection is prohibited without written permission, then block named AI crawlers and every unnamed bot.
  • TikTok names more than 20 AI crawlers and fetchers in a single block, including OAI-SearchBot, ChatGPT-User, Claude-SearchBot and Perplexity-User. Its general rule (*) allows pages such as /discover, /tag and /music, which is the rule Googlebot follows.
  • X names only a few bots. Googlebot and Bingbot share one set of rules; Google-Extended and Meta’s crawlers are blocked; everyone else falls under Disallow: /.
  • LinkedIn is the most nuanced. It blocks training crawlers (GPTBot, ClaudeBot, Google-Extended) and live fetchers (ChatGPT-User, Claude-User, Perplexity-User), but gives OAI-SearchBot and Claude-SearchBot a long list of rules similar to Googlebot’s. The file notes that crawling LinkedIn requires permission.

Robots.txt files change without notice. Check the links in our sources for the current version.

What still gets through

It would be wrong to say AI can see nothing on social media.

  • Google sees a lot. Googlebot is allowed on all five platforms, so public pages and profiles can be in Google’s index. Google’s AI Overviews and AI Mode build on that index, so a public social page can be a source there.
  • Live fetches can ignore robots.txt. OpenAI says ChatGPT-User is triggered by users and robots.txt rules “may not apply”; Perplexity says its user fetcher “generally ignores robots.txt rules.” But they still face login walls and bot checks.
  • Other websites quote social content. News sites, directories and blogs that embed or describe your posts can be read and cited.

Logins and bot checks: what a crawler actually gets

We also made a simple test: one plain request (no login, no JavaScript) to a public profile on each platform, on 27 October 2026. This isn’t how every crawler works, but it shows what a basic fetcher gets.

Platform What came back
Facebook (public brand page) An error page: “Sorry, something went wrong”
Instagram (public profile) A page shell with no readable text
TikTok (public profile) A “Please wait…” check page
X (public profile) Name, bio, follower count and “Log in or sign up” prompts
LinkedIn (public company page) Readable company information, with sign-in prompts

And some content is private by design: posts shared with friends only, private groups, Stories, Messenger and LINE chats. No crawler, AI or otherwise, can see those.

What this means for your business

If your business information lives only on social media, AI assistants outside Google have very little they can read and cite. Even with Google, you don’t control what it picks up from a social page.

Put the facts AI assistants need — services, your prices, location, hours, FAQs — on your own website, in plain text. Use social media to reach people, and link back to the site.

Keep your social profiles complete and consistent with your website: same name, address, phone and description. That consistency helps AI connect the dots. But make your website the main source.

We explain the full case in why your website matters more than social media in AI search. For what to put on the site, use our AI-ready website checklist, and for the wider picture, read our AI SEO guide.

If you need a website that AI can read, we design and build it — usually live in about a week, managed afterwards in our customer portal FoxWorks. If you’d like help getting your brand cited in AI answers, see our AI search & SEO service.

FAQ

Can ChatGPT see my Facebook posts?

Usually not. Facebook blocks OpenAI's crawlers in robots.txt, and anything not shared publicly needs a login. ChatGPT may know about your business from other sources, such as your website or directory listings.

Can Google AI Overviews use my Instagram profile?

Possibly. Instagram allows Googlebot to crawl most public pages, so a public profile can be in Google's index, which AI Overviews and AI Mode use. Gemini grounding is a different case: Instagram blocks the Google-Extended token.

Why does LinkedIn allow some AI crawlers?

LinkedIn's robots.txt gives OAI-SearchBot and Claude-SearchBot similar access to Googlebot, with many paths excluded, while blocking training crawlers such as GPTBot and ClaudeBot. Platforms set these rules themselves and can change them at any time.

Will this change?

It can. robots.txt files are updated without notice. We'll recheck and update the date on this page.

Sources

[Author name][Role and experience]