Yannis Antonoglou

It was like a startup at the time.

I was like one of the founding engineers.

And then I did a lot of the deep reinforcement learning research that came out of DeepMind while I was there.

I worked on DQN, AlphaGo, AlphaZero, NuZero, and also did RLHF for Gemini before I left.

So I actually have like a lot of

kind of like experience and a lot of, you know, I've spent most of my career doing research and just like really trying to push the boundaries of like the frontier and like, you know, just ensuring that we have like the most powerful algorithms for reinforcement learning and AI in general.

And one of the things that, you know, I think that like everyone who's done a lot of research can agree on is that

Open science, publications, sharing your findings, uploading your code is something that is the only way to really ensure that the progress is actually accessible by everyone and also that we can speed up the progress of intelligence and of science as a whole.

Yeah, that's a really good question.

The most important thing to note about open models is that they're especially useful to people who want to build with them or they want to just do more research or they want to just tinker with the model and just try and build their own solutions.

And a lot of the innovations happen because people are curious and they want to just take models or they want to just take a...

for example, an operating system.

So in the past, you had like the Ubuntu and the Linux operating system and people just like, you know, try and fix bugs and just like make it more reliable and like more secure.

At the same time, they'll just like try to contribute and at the same time, it could be used as an educational tool for like the next generation of computer scientists to learn how to program operating systems and how they work and, you know, do their assignments or just like, you know, find better solutions to existing problems.

So,

Open weights and open source is the way to ensure that innovation happens at scale and at the same time you have the new generation of scientists actually really understanding how these systems work and being able to contribute to the next iteration and generation of systems.

That's awesome.

Yeah, that's straight to the point, I guess.

So,

Yeah, so what we're building is that we want to build the most powerful agenda models in the world and make them open weight and accessible to everyone.