GoogleOther
Google's generic internal crawler. Google does not document a training purpose for it - despite common claims.
Reference reviewed by Sona
| Operator | |
|---|---|
| Powers | Internal Google product-team crawls |
| Purpose | General crawl |
| User-agent token | GoogleOther |
| Respects robots.txt | Yes |
GoogleOther is described by Google as a "generic crawler that may be used by various product teams for fetching publicly accessible content." It exists so internal one-off and R&D fetches do not have to masquerade as Googlebot and do not consume Googlebot's crawl budget.
It is worth being careful about what Google actually claims here, because this token is widely described online as Google's AI training crawler. Google's documentation says no such thing - it says generic, and it names no product. AI teams at Google plausibly sit within "various product teams", but the training-crawler framing is inference, not documentation. Google's actual training control is the separate Google-Extended token.
GoogleOther respects robots.txt rules written for its own token, and blocking it does not affect Googlebot, Search rankings, or AI Overviews eligibility. It also has GoogleOther-Image and GoogleOther-Video variants that fall back to this group unless given their own rules.
Because the purpose is undeclared and the traffic is usually light, this is a low-stakes token in both directions. Blocking it is a reasonable default for anyone who wants Google's access scoped strictly to Search; allowing it costs little.
How GoogleOther behaves
- Google declares no specific product or purpose for it beyond generic product-team fetching.
- Exists partly so internal crawls do not consume Googlebot's crawl budget.
- GoogleOther-Image and GoogleOther-Video fall back to this group unless separately specified.
- Blocking it has no effect on Googlebot, Search rankings, or AI Overviews.
Full user-agent string
Mozilla/5.0 AppleWebKit/537.36 (KHTML, like Gecko; compatible; GoogleOther)
Allow GoogleOther
Internal Google crawls can fetch your content without you giving up any Search control, and volume is typically light.
User-agent: GoogleOther Allow: /
Block GoogleOther
You want Google's access limited strictly to Search indexing - and since the purpose is undeclared, declining it is defensible.
User-agent: GoogleOther Disallow: /
How to verify GoogleOther
Reverse DNS + published IP ranges
GoogleOther is covered by Google's common-crawlers IP range feed rather than the Googlebot-specific file, and reverse DNS resolves under googlebot.com or google.com like Google's other infrastructure. Use forward-confirmed reverse DNS - reverse-resolve, then forward-resolve the hostname and check it returns the original address.
Documented reverse-DNS suffixes: .googlebot.com, .google.com
Commonly confused with GoogleOther
Google-Extended is the documented Gemini training control. GoogleOther is often mistaken for it, but Google claims no training purpose for GoogleOther at all.
Separate tokens and separate crawl budgets. Blocking GoogleOther does not affect Search; blocking Googlebot does.
GoogleOther FAQs
Is GoogleOther Google's AI training crawler?
Not according to Google. The documentation calls it a generic crawler used by various product teams and names no purpose. The widespread claim that it is a training crawler is inference. Google's documented training control is the Google-Extended token.
Does blocking GoogleOther affect my Google Search rankings?
No. It is a separate token with its own robots.txt group and its own crawl budget. Blocking it leaves Googlebot, your rankings, and AI Overviews eligibility untouched.
Why does GoogleOther exist at all?
So internal, one-off, and R&D fetches by Google product teams are identifiable as such and do not consume the crawl budget allocated to Googlebot for Search indexing.
Do I need separate rules for GoogleOther-Image and GoogleOther-Video?
Only if you want different treatment for them. Both fall back to the GoogleOther group when they have no rules of their own.
Can GoogleOther read your page right now?
Test any URL and see exactly what AI crawlers receive.