Model training · Google
Google-Extended
A robots.txt control token, not a separate crawler: Googlebot does the fetching. Disallowing Google-Extended stops your content being used to train Gemini models and to ground Gemini answers. It does not affect Google Search, AI Overviews or AI Mode.
- Operator
- Purpose
- Model training
- Used for
- Gemini training and grounding
- Obeys robots.txt
- Yes
- robots.txt token
Google-Extended- User-agent string
- Not published by the operator
- IP list
- Not published
- Official docs
- developers.google.com
- Feeds
- Gemini
Block Google-Extended in robots.txt
User-agent: Google-Extended
Disallow: /Well-known sites that block Google-Extended
facebook.com #4, fbcdn.net #7, instagram.com #14, amazon.com #30, whatsapp.net #42, whatsapp.com #66, yahoo.com #72, msn.com #92, chatgpt.com #112, snapchat.com #157, tumblr.com #160, nytimes.com #163, gravatar.com #165, bbc.com #216, sciencedirect.com #224, bbc.co.uk #236, launchpad.net #281, nature.com #297, alibaba.com #310, amazon.co.uk #326, espn.com #363, bloomberg.com #386, amazon.de #387, go.com #403
Other Google bots
GoogleOther (mixed / general) · Google-CloudVertexBot (ai agent) · Google-Agent (ai agent) · Googlebot (ai search index)
Questions
What is Google-Extended?
Google-Extended is a model training bot operated by Google, used for Gemini training and grounding. A robots.txt control token, not a separate crawler: Googlebot does the fetching. Disallowing Google-Extended stops your content being used to train Gemini models and to ground Gemini answers. It does not affect Google Search, AI Overviews or AI Mode.
How do I block Google-Extended?
Add "User-agent: Google-Extended" followed by "Disallow: /" to the robots.txt file at the root of each host you want to protect.
Does Google-Extended respect robots.txt?
Yes. Google documents that Google-Extended honours robots.txt directives.
How many websites block Google-Extended?
In a Oct 29, 2026 analysis of the top 10,000 websites by Kuldeep Paul, 15.5% of sites with a robots.txt file fully disallowed Google-Extended.
Analysis using this data
- Who blocks AI crawlers? What 10,000 robots.txt files revealPublishers block training, not answers; half of all blocklists leak; Cloudflare writes the most common AI policy.
- The opt-out asymmetry: you can block the small AI engines, not the big oneThe AI surfaces with the most users are the ones publishers can't refuse and can't measure.
How to cite this data
This dataset is free to use, including commercially, under Creative Commons Attribution 4.0. The one rule is attribution:
- Credit the source as “Kuldeep Paul, kuldeep.cc” wherever you use a figure, chart or table from it.
- Link to this page (https://kuldeep.cc/data/ai-search/crawlers/google-extended), not to the download file, so readers can check the source of every number.
- Include the date. These figures change; say “as of Oct 29, 2026” or link to the version you used.
- Mark changes. If you filter, recalculate or chart the data yourself, say so (“analysis based on data from…”).
- Cite the original source for single figures when you quote one reported number; it is linked in every row.
In running text: according to Kuldeep Paul's Google-Extended — AI crawler directory (kuldeep.cc). More in the methodology.
Paul, K. (2026). Google-Extended — AI crawler directory [Data set]. kuldeep.cc. https://kuldeep.cc/data/ai-search/crawlers/google-extended
Paul, Kuldeep. "Google-Extended — AI crawler directory." kuldeep.cc, 29 October 2026, https://kuldeep.cc/data/ai-search/crawlers/google-extended.
Paul, Kuldeep. 2026. "Google-Extended — AI crawler directory." Dataset. kuldeep.cc. Updated Oct 29, 2026. https://kuldeep.cc/data/ai-search/crawlers/google-extended.
@misc{paul2026googleextended, author = {Paul, Kuldeep}, title = {Google-Extended — AI crawler directory}, year = {2026}, howpublished = {\url{https://kuldeep.cc/data/ai-search/crawlers/google-extended}}, note = {Dataset, CC BY 4.0, updated 2026-10-29} }
<a href="https://kuldeep.cc/data/ai-search/crawlers/google-extended">Google-Extended — AI crawler directory</a> (data: Kuldeep Paul, kuldeep.cc)