Richard Socher

So I think that that is the ambition.

We have a singular focus, right?

There's a lot of labs that do a lot of different things.

For recursive, that is like recursive superintelligence in the name.

That is what we're doing.

The second one is infrastructure.

We're like custom building everything from the ground up.

with superintelligence in mind.

That means that we design sort of all the teams, the resources, the infrastructure, the architecture, and so on from the ground up for autonomous scientific discovery.

So that's number two is infrastructure.

Number three is that we're really inspired by so-called open-ended systems.

It's a fundamentally different approach that relieves us of some of these constraints that others face.

And why is the early view constraints?

So in many ways, like what a lot of AIs and AI labs are doing is they kind of define these spiky intelligences.

They're like, this is the particular data set, benchmark or so on that we want the AI to be really good at.

And then the AI gets really good along that one line.

but then there are all these sort of valleys between the spikes.

And so we believe that instead of optimizing one fixed target that will ultimately plateau, we believe in open-endedness and you have interestingly different sets of ideas and they can compete with one another.

And they keep improving without this sort of obvious ceiling of like, oh, you got 100 out of 100 questions right on this kind of test.

And then of course it's the team.