Adam Elga

Maybe the AI shouldn't trust us.

What in the world?

Well, that's tricky.

Look at what happens if you think the AI is in somewhat of the undermining, self-undermining type situation that a Boltzmann brain is in or the x-ray machine looking at itself is in.

If you're in that situation, you think, I can't trust anything.

Now you start doubting.

rewinding to some very, very cautious prior of I don't know what's going on.

Think about how dangerous that is if a creature with that kind of standpoint is in power.

Pleasure to chat.

Thank you.