Paste a suspicious email. You get a straight answer on whether the sender is who they claim to be, and the reason why.
No email yet? Try a real one: , ,
Free. It runs in your browser, so nothing is sent anywhere or stored.
It compares the sending address against a built-in list of 88 real company domains, in your browser. It looks at where an email came from, not how well it is written.
The problem
AI writes a flawless, personalised pitch and puts a real person's name on it. Your spam filter waves it through because the words look clean. The big-company filters do not cover you. You're on your own, until you check who actually sent it.
Glossy role, your name on it. Reply-to: a throwaway domain that isn't Dentsu.
A real broadcaster's name and photo. Sent from a Gmail address. The BBC never books from Gmail.
A real group's name, borrowed. A look-alike agency domain one letter off the real firm.
How it works
Trying to spot "AI-written text" is guesswork, and it is getting worse. Proof of Sender ignores the writing and checks the things that are hardest to fake.
Does the address belong to the company named in the email? Or is it one letter off, or a free Gmail account?
Is the named person, role or group real and checkable, or a well-known name borrowed to borrow trust?
If the same message reached more than one of your addresses, that is a list, not a person. Compare the two and the wording matches.
What a verdict looks like
A score is never shown on its own. This is a real fake that reached my inbox, and the actual output the checker gives for it.
Who it's for