Rob Wiblin

I guess there's all of these examples of the AIs doing crazy stuff, but

You can always say, well, it was just playing a role, for example.

And I guess for you and me, we're like, yeah, but it might end up playing a role.

That's how it could end up doing bad stuff.

Or this is a demonstration of other failure modes that we'll see later on, that just in general, we don't have a full grip on it.

I suppose it's so hard.

to get people to have to believe something that they really don't want to believe.

Or that seems incredible to them.

Yeah, so I think that's like real research.

Yes.

Apart from Scientist AI and this, are there any other top requests that you have of people in the companies?

Or is there any common practice that you think is particularly crazy that they should maybe cut out?

Yes.

Yeah, I think we're currently on track to start on fully automated AI R&D and have the companies be saying, well, we got the AI to monitor itself and it didn't flag anything.

And that's why we feel pretty good about this.

I actually think that that is like the most likely outcome.

And I guess we'll see how that goes.

Fingers crossed we can do better.

Earlier on you were talking about, as you've become more optimistic in a sense that we do at least in principle have a solution to the control problem, you become more worried about the human concentration of power stuff.

Do you have any suggestions or any policy ideas here?