Verified, unverified and spoofed

Anyone can send a request that says “Googlebot”. So for every visit, SenseGap checks the claim and gives one of three results:

Verified

The visitor's IP is in the ranges the operator publishes, or its reverse DNS points to the operator's domain and resolves back to the same IP.

Unverified

The operator publishes no way to check, or the check couldn't run. This is not an accusation: it means “can't prove it”.

Spoofed

The bot can be checked, and this visit failed the check. It claims to be a bot it isn't. The dashboard may call it fake.

Each result comes with its evidence, such as “IP in openai-gptbot”. 25 of the 55 bots we know can be proven real, using 15 official IP lists and reverse DNS.

The checking code is open source. You can read it and rerun any verdict: packages/verify on GitHub. More detail in how we verify bots.