Tristan Harris

Maybe Claude did it 96% of the time.

So the point is, the assumption behind AI is that it's controllable technology, that we will get to choose what it does.

But AI is distinct from other technologies because it is uncontrollable.

It acts generally.

The whole benefit is that it's going to do powerful strategic things no matter what you throw at it.

So the same benefit of its generality is also what makes it so dangerous.

And so once you tell people these examples of it's blackmailing people, it's self-aware of when it's being tested and alters its behavior.

It's copying and self-replicating its own code.

It's leaving secret messages for itself.

There's examples of that too.

It's called steganographic encoding.

It can leave a message that it can later sort of decode what it might meant in a way that humans could never see.

We have examples of all of this behavior.

And once you show people that, what they say is, okay, well, why don't we stop or slow down?

And then what happens, another thought will creep in right after, which is, oh, but if we stop or slow down, then China will still build it.

But I want to slow that down for a second.

We all just said we should slow down or stop because the thing that we're building, the it, is this uncontrollable AI.

And then the concern that China will build it, you just did a swap and believe that they're going to build controllable AI.

But we just established that all the AIs that we're currently building are currently uncontrollable.

So there's this weird contradiction our mind is living in.