Nick Heiner

You know, we're working on a bunch more that we haven't released publicly, but that, you know, we've just we've just deployed to clients.

Really, what we're looking at in terms of how we prioritize our work with clients is.

looking at the intersection of economically valuable work for AIs to be doing and places where it could be deployed reasonably easily.

So, you know, you could imagine medical environments, you know, we're definitely doing some work in that space, but there are just a lot of regulatory barriers to deploying really any sort of technology into health care.

So, you know, if you have something else like, you know, finance where those firms are like super hungry and are, you know, willing to be adventurous to generate alpha, like that is sort of a great intersection of things for us to target.

Unfortunately, contractually, we're not allowed to talk about specific customers, but I'll say, you know, we work with the Frontier Labs and we work with other labs that are not Frontier Labs.

Yeah, so one thing that's really interesting about training AIs, and you'll see this also in the finance report that we did, where we gave three models, Gemini, Claude, and JGPT, like professional finance tasks, is that with humans, generally, if someone can do some really deep analysis, they can also make a PowerPoint.

But with models, it's not necessarily the case.

Like you sort of see failures all up and down the stack.

And so you'll sort of have like, you know, a model like do an amazing JavaScript calculation and then just like not make the slides.

So that's been one interesting thing is like,

It's not really the case that there's sort of one area and you like solve that area and then it's done forever.

Like instruction following, you know, you just dump a bunch of data into instruction following and then you never need to think about it again.

Or personality or, you know, being grounded in the factual materials presented.

Like that's part of why we're so bullish on RL environments is that it really exercises all that stuff at the same time.

Because if you make a mistake in any one of those dimensions, you're going to fail the task.

And so user protect against regressions as opposed to older training regimes that would like really focus on just one of those slices.

So yeah, in terms of what's the hardest to train the model?

The hardest things to train are the things that are hardest to verify.

So the reason that models are really good at math and code is because there are a lot of verifiable outcomes.