Start here
The free tool below runs entirely in this browser — no account, nothing stored or transmitted. Redact anything you would not want kept before you type it.
Open the companion tool →Why bot-detection is a trap
Trying to detect a bot by its voice is a losing game — generated speech now pauses, backchannels, and even interrupts. Worse, it does not matter: a real human scammer is just as dangerous as a bot, and the same checks stop both.
The checks that matter are about the ask and the identity claim: what does the call want you to do, and can it verify its identity on a channel you initiate? Those work on bots, humans, and everything in between.
The protocol
- Describe the call: who it claimed to be and what it asked you to do.
- Read the disambiguation basis: the ask and the identity claim each get a line.
- Do not comply on the call. Hang up and verify on a channel you initiate.
- Report unwanted robocalls to the FCC and the FTC; block the number after.
What the tool does
Describe a suspicious call and get a disambiguation reading with the ask and identity claim broken out. It runs in this browser; nothing is sent or stored.
Limit first: This check can raise your confidence, but a clean result doesn't prove a call is real or safe. It structures the ask and identity questions — it does not record or analyze the call.
Open the companion tool →One human next step
Hang up and verify on a channel you initiate. Report unwanted robocalls at the FCC's consumer page and the FTC's Do Not Call reporting.
Research log and safety checks
AI-assist path: not used for a verdict. This artifact is an inspectable disambiguation aid; a remote model would add an opaque bot-detection claim where the reader needs the ask-and-identity basis. §6.10: not a frontier-pace topic.
§0.5 protection result: category is deepfake-media-verification; crisis-adjacent: no; fear/urgency/scarcity toolkit used: no. The protective function is free and the paid feature is convenience only.