Nick Heiner

Yeah.

Like did the code run?

Yeah.

But then imagine something like a model writing a textbook.

How are you going to verify the quality of a textbook?

Are you going to have a person read 300 pages?

And even then, it's like, okay, well, maybe you didn't like the textbook, but when students actually use it, they go out into the real world and they're much more proficient at the skill.

Like, are you going to run that experiment?

You know, are you, are you going to generate a thousand model trials of running textbooks and then teach 10,000 different students?

Like, no, like obviously it's not going to scale.

So like, there's a really deep research problem we're working on, which is how do you verify increasingly open-ended problems in a way?

Yeah, yeah, exactly.

In a way that's like stable and is tracking the real world value that we care about.

Yeah.

And, you know, you can even, I mean, humans have the same problem with training.

Like, again, to bring it back to VCs.

You make a bunch of investments and like funds are 10 years long for a reason.

Like you actually don't know if you've made good investments for years.

And so that's a really hard feedback signal to learn from.

Right.