There's no way I should be bound by terms of service that are secret unless I know the "code-phrase" with which to ask for it. Google's whole point was to deceive the user into thinking it isn't a bot -- so you wouldn't even know to ask it.

People will be forced to "discover" that they've been lied to, when the bot is caught in a loop and they're going through emotional distress thinking the "person" on the other end is having some sort of mental breakdown.

It's illegal in many states to record a conversation without obtaining consent which is something this product does as a matter of course. This product shouldn't be legal to operate in those states, so you don't have to look very far before finding critical flaws in this approach.

That said, "identify on challenge" is something that Google might be more likely to adopt rather than "identify on call pickup". The latter being something they might spend more lobbyist dollars to oppose.

But what you've said is certainly (or should be) a real concern.

I agree, but the social pressure not to use it would be immense. Imagine the awkward moments where we start confronting people who call us and asking them if they're real.

It's already happening. I'm getting robo calls from some police donation fund that will slow down and restart speaking if you interrupt them (took me a few moments to realize it was a recording).

Charities and political organizations are excluded unfortunately from robocall regulation. Google, though, is not.

If I were to start receiving Duplex calls, and could detect it, I would report each one to the FTC.

Perhaps they should identify themselves upfront? "Hello, this is an automated call on behalf of..."

This reminds me of some of Isaac Asimov's short stories in which the Three Laws are used to force robots to identify themselves.

Recording calls without both parties' knowledge can be illegal.

I think at the very least the bot should disclose that it is a bot right from the beginning.