Google-Extended: Google's AI Training Opt-Out Token
Google · Robots.txt control token (not a crawler) · Last verified October 2026
Google-Extended is not a crawler — it's a standalone robots.txt token controlling whether Google may use your content to improve its AI models (Gemini) and to ground answers in the Gemini app and Vertex AI. It never fetches anything and never appears in your logs; Googlebot does the crawling. Blocking it does not affect Google Search rankings or AI Overviews — but it can keep your content out of Gemini's grounded answers.
What is Google-Extended?
A standalone product token, not a bot. It controls whether Google may use your content to improve its AI models (Gemini). It performs zero fetches.
Why there is no user-agent string
Google-Extended never fetches anything, so it has no user-agent string. You'll never see it in your server logs — the crawling is done by Googlebot. The token exists only as a name you can reference in robots.txt.
Does Google-Extended respect robots.txt?
Not applicable — it's a token, not a bot. But Google honors the token: a User-agent: Google-Extended + Disallow: / rule opts your content out of Gemini training and grounding use.
How to use the Google-Extended token in robots.txt
To opt out of training use, add this to robots.txt. (Allowing is the default — you only need a rule to block.)
User-agent: Google-Extended Disallow: /
What blocking Google-Extended does (and doesn't do)
Blocking Google-Extended opts you out of Google's AI training data use and out of grounding in the Gemini app and Vertex AI, period. It does not remove you from Search. It does not stop AI Overviews — those use Googlebot. This is the single most misunderstood token in AI SEO, and many sites block it accidentally via copied templates. If your goal is appearing in AI Overviews, this token is irrelevant — check Googlebot instead. If you want the Gemini app citing your pages, think twice before blocking it.
Beyond robots.txt
robots.txt is a request, not a barrier. For server- and CDN-level blocking (nginx, Apache, Cloudflare WAF), see blocking beyond robots.txt on the overview page.
Google-Extended vs. related bots
Changelog
Introduced by Google as the dedicated training opt-out token, separate from Googlebot. Last verified October 2026.
What does Google-Extended see on your site?
Operator docs: Google · All AI crawlers