Google-Extended is a robots.txt token, not a crawler. Google reads it to decide whether your content may be used for its AI — it is not an agent that fetches pages, and it is not Googlebot. The question it answers is narrow: may Google's AI use what its search crawler already collected? Blocking it does not remove you from Google Search, and allowing it does not change your ranking.
A User-agent: Google-Extended group with Disallow: / tells Google not to use your content for AI. Your pages stay in Google Search, your local-pack presence is untouched, and Googlebot keeps crawling as before — search indexing and AI usage are controlled by different names. What you give up is the possibility of your content informing Google's AI answers, which for a local business is a real channel: an assistant that may not use your content cannot cite your business from it.
Name it in its own robots.txt group — a crawler-side reader obeys the most specific group naming it, so a wildcard allow elsewhere in the file does not override a Google-Extended block, and vice versa. Our own robots.txt allows it explicitly, alongside GPTBot, OAI-SearchBot, ClaudeBot and PerplexityBot, because being readable to assistants is this site's business. Whichever you choose, choose it deliberately: our free scan reports how your robots.txt treats all six AI agents, by name, so the setting you think you have is the one you actually have.
No. It controls AI usage of your content, not search indexing — Googlebot and your ranking are unaffected. The two are controlled by different robots.txt names precisely so you can decide them separately.
It is Google's control for AI use of your content. Gemini reads through Google's renderer rather than fetching separately, so Google-Extended is the lever Google gives you for the AI side while search stays governed by Googlebot.
Usually not. 45% of consumers now use AI to find local businesses (BrightLocal, 2026); for most owners the trade — AI answers may use your public business facts, in exchange for being citable — favours allowing it. If your content is your product, the calculus can differ.
Read yourdomain.com/robots.txt and look for a group naming Google-Extended, remembering that the most specific group wins. Or run the free scan, which parses the file with the crawlers' own precedence rules and names what is blocked.
Where does your site stand? The scan is free, takes about 60 seconds, and needs no account. It fetches your homepage exactly the way GPTBot, ClaudeBot and PerplexityBot do and tells you what they receive.