Google-Extended

What does Google-Extended control?#

Google-Extended controls whether content Google crawls from your site may be used to train future Gemini models and to ground answers in the Gemini app and Vertex AI; it does not affect Google Search, AI Overviews or AI Mode, and Google says it isn’t a ranking signal.

Google’s crawler documentation calls it “a standalone product token” for managing whether content “may be used for training future generations of Gemini models that power Gemini Apps and Vertex AI API for Gemini and for grounding in Gemini Apps and Grounding with Google Search on Vertex AI.” The same page is explicit about Search: “Google-Extended does not impact a site’s inclusion in Google Search nor is it used as a ranking signal in Google Search.”

Grounding is the part people miss. The Gemini app answers current questions by retrieving pages from Google’s index, and a page behind a Google-Extended block can’t be one of them. See grounding and the Gemini SEO guide.

Does blocking Google-Extended remove you from AI Overviews?#

Blocking Google-Extended does not remove you from AI Overviews or AI Mode, because Google’s AI features in Search are governed by Googlebot; to leave them, use Search Console’s Search generative AI setting, nosnippet or noindex.

Google’s AI features documentation explains why: “AI is built into Search and integral to how Search functions, which is why robots.txt directives for Googlebot is the control.” Blocking Googlebot would take you out of Search entirely, so the narrower switches below are the ones to use.

The two AI opt-outs are mirror images. The Search Console control, live worldwide since 31 August 2026, removes a property from AI Overviews and AI Mode but not the Gemini app; Google-Extended removes you from the Gemini app but not AI Overviews. See AI Overviews and snippet controls.

Google-Extended vs Googlebot: what's the difference?#

Google-Extended is only a robots.txt token with no user agent of its own, while Googlebot is the crawler that actually fetches your pages for Search and its AI features; Google-Extended just tells Google how content Googlebot already crawled may be used for Gemini.

Google says so directly: “Google-Extended doesn’t have a separate HTTP request user agent string. Crawling is done with existing Google user agent strings; the robots.txt user-agent token is used in a control capacity.” Two practical consequences follow. You will never see Google-Extended in your server logs. And blocking it doesn’t reduce crawl load, because Googlebot keeps visiting either way — the token only changes what Google may do with what it fetched.

Should I block Google-Extended?#

Block Google-Extended only if keeping your content out of Gemini training matters more than being cited in the Gemini app, because Google ties the two together: one token opts you out of both, while Search, AI Overviews and AI Mode are unaffected either way.

The Gemini app had 950 million monthly users in the second quarter of 2026, per Alphabet, and studies find it cites only about three sources per answer, so each citation is scarce. For a business that wants to be recommended, a Google-Extended block trades those citations for a training opt-out. For a publisher whose content is the product, that trade can be worth it.

# Search, AI Overviews and AI Mode
User-agent: Googlebot
Allow: /
 
# Opts out of Gemini-app grounding AND Gemini training.
# Leave this out if you want to be cited by Gemini.
User-agent: Google-Extended
Disallow: /

How do I check whether Google-Extended is working?#

Google-Extended can’t be checked in server logs because it never visits — Googlebot does the crawling — so the check is reading your robots.txt the way Google does: find the group the Google-Extended token matches on every host, and confirm it allows or disallows what you intend.

1. Open /robots.txt on every host, including www, the bare domain and each subdomain. Rules apply only to the host that serves them.
2. Find the group that matches Google-Extended: its own group if one exists, otherwise User-agent: *. Google never combines a named group with the * group.
3. Read the file as served, not as committed. Cloudflare’s managed robots.txt prepends rules to your file, including a disallow for Google-Extended — which opts you out of Gemini-app grounding, not just training.
4. Check the effect where it lands. Run your key buyer questions in the Gemini app and see whether your pages appear among its sources. That’s the only visible result of the token.

Changes aren’t instant: Google generally caches robots.txt for up to 24 hours. And because the token covers only future training and grounding, it won’t pull your pages out of models trained before the change. See robots.txt for how groups are matched.

Related terms#

• AI search & GEOAI OverviewsAI-generated summaries shown at the top of some Google search results, written by a Gemini model from pages in Google’s search index and linked to their sources — so they are won through ordinary Google eligibility, not a separate submission or special markup.Read the entry
• AI crawlersrobots.txtA plain-text file at a site’s root that tells crawlers which paths they may fetch, rule by rule for each user agent — a voluntary standard that reputable search and AI bots follow, which controls crawling but not whether a URL gets indexed.Read the entry
• Technical SEOSnippet controlsRobots directives and HTML attributes — nosnippet, max-snippet and data-nosnippet — that limit how much of a page search engines may quote in results, and in Google they also limit what AI Overviews and AI Mode can use.Read the entry
• How LLMs answerGroundingThe process of tying an AI model’s answer to specific, verifiable sources — usually web pages retrieved at the moment of the question — so the claims in the response can be checked and cited, rather than generated from the model’s memory alone.Read the entry
• AI crawlersAI crawlersAutomated bots run by AI companies that fetch web pages for one of three jobs — training models, building an AI search index, or retrieving a page live for a user’s question — and because each job uses its own user agent, each can be allowed or blocked separately.Read the entry
• GSCMeasurementGoogle Search ConsoleGoogle’s free tool for site owners that reports how a site performs in Google Search — queries, impressions, clicks, position and indexing — and since 2026 it also reports impressions in Google’s generative AI features and lets owners opt out of them.Read the entry