CEOInterviews.AI
Start App
Try: AI strategy tariffs earnings guidance layoffs interest rates growth

Oriol Vinyals

Co-Founder, Discovery Loop

Search every verified Oriol Vinyals interview, podcast appearance, and on-the-record quote — each transcript cross-checked by AI and human review to confirm speaker identity. Oriol Vinyals, co-founder of Discovery Loop, gave a talk at the University of California, Berkeley, in August 2026, discussing his perspective on recursive self-improvement (RSI) in artificial intelligence. Drawing on his background at DeepMind, he contrasted traditional reinforcement learning loops with the concept of an agent modifying its own environment, harness, or model weights. He noted that while coding and experimentation have seen significant progress, evaluation and ideation remain difficult components that may slow the pace of fully automated RSI. Vinyals also cautioned that the impact of RSI could be limited by physical constraints, such as chip speed and data transmission, and that some human performance may already be near an asymptote, meaning further improvement might not yield drastic gains. He suggested that RSI could increase the speed of progress in engineering and science tasks by tenfold or more, but emphasized the need for careful consideration of goal-setting, misalignment, and resource reallocation as research labs adapt to these changes.

Recent appearances

Notable quotes

“It's definitely clear that RSI will increase perhaps 10x or more the speed of progress of certain in engineering and science task for sure. But you know, be mindful of how we set the goals, misalignment, resource reallocation, how the labs are run might change. So we need to be quite careful about how to take different steps in this. And as I was arguing, it might not come as fast as we thought it would because first”

“The way to think about it is that up until now our data center business has been almost entirely GPUs. Vera is our CPU, and you know what's really exciting is that Vera opens a brand new 200 billion TAM for NVIDIA, a market we have never addressed before, and every major hyperscaler is partnering with us to get it deployed, and that's huge for the company because for the first time we have a second engine of that sca”

“The way to think about it is that up until now our data center business has been almost entirely GPUs. Vera is our CPU, and you know what's really exciting is that Vera opens a brand new 200 billion TAM for NVIDIA, a market we have never addressed before, and every major hyperscaler is partnering with us to get it deployed, and that's huge for the company because for the first time we have a second engine of that sca”

“The problem is, random policy will not do much in this game that's useful. And so in a way, the bitter lesson discussion, which is, should you gather as much imitation learning experience to bootstrap the knowledge about the domain even in a game like StarCraft, which arguably, is not AGI complete, as complex as it is, didn't work to the best of our abilities. So for us, it was very clear we would need an imitation l”

“If 10 years ago, 5 years ago even, I would have been given the models today, and I would say, look, there's a secret lab, this is a model, play with it and tell me if you think this is actually close to a general intelligence; I would have claimed, oh, yeah, that comes from a future where AGI basically either has happened or I can see that this is very close to it. So the closer you are, the more you find, oh, but it”

The summary and quote tags on this profile are produced with AI assistance from verified, first-person interview transcripts, then checked by our team to confirm the speaker's identity and the accuracy of every quote. See how we verify →