Max Tegmark

And this is an idea I give credit to Steve on 104.

So that it will only run the code if it can prove... Instead of not running it, if it can prove that it's not trustworthy, if it will only run it if it can prove that it's trustworthy.

So it asks the code, prove to me that you're going to do what you say you're going to do.

And it gives you this proof.

And you, a little proof checker, can check it.

Now you can actually trust...

an AI that's much more intelligent than you are, right?

Because it's problem to come up with this proof that you could never have found, but you should trust it.

I have to lie to you and give me a proof that I'm going to think is correct?

But it's not me who's lying to you.

It's the trick, my proof checker.

It's a piece of code.

Yeah, I don't buy that argument.

I think no matter how...

super intelligent an AI is, it's never gonna be able to prove to me that there are only finitely many primes, for example.

It just can't.

And it can try to snow me by making up all sorts of new weird rules of deduction that say, trust me, the way your proof checker works is too limited and we have this new hypermath and it's true.

But then I would just take the attitude, okay, I'm gonna forfeit some of these

the supposedly super cool technologies.

I'm only gonna go with the ones that I can prove in my own trusted proof checker.