Ezra Klein pressed Huang on lab researchers who say they are losing the ability to evaluate their own models. Huang's answer was that most labs today put roughly 80% of their effort into capability and 20% into safety and verification, and that the ratio should flip — at NVIDIA, he said, 80% goes to verification. He compared it to wishing the car industry had reached today's braking technology ninety-nine years sooner.
I don't believe that. I believe that their researchers are working every single day to learn about how to evaluate these systems. Verification, so you know, 10%, 20% of our company is dedicated to design, 80% is dedicated to verification. Today, most labs, understandably, is 80% dedicated to capability and 20% dedicated to safety, verification, eval.
This is the flip. This transition you're talking about. That's right.
That's right. AI needs to accelerate to be safe. I want them to get more compute, but allocated towards evaluation, to alignment. And I think they're doing that. If I were in the car industry 100 years ago, I would rather the car industry accelerated to today in one year because I believe today's car is way more safe than a car 99 years ago. And ABS technology, automatic breaking, requires computer vision technology, sensor fusion technology, radars and cameras, and all that technology coming together in order to break when you should and not break when you shouldn't. That technology is extremely hard. I would have hoped, everybody would have hoped that ABS technology existed 99 years ago. A lot fewer children would have been killed. And so. You know, airbags, safe, you know, seatbelts. I mean, all of that stuff, self-tightening seatbelts, all of that stuff. Could you imagine? That's all technology. Accelerate the living daylights out of that development. And so when I say we need to accelerate AI technology, people think for some reason safety is not part of that. Safety is part of it. Alignment is part of it. Eval is part of it. Guardrailing, sandboxing, the isolation technology, monitoring technology, telemetry technology, external AI monitor technology, all of that stuff is AI technology. Accelerate the living daylights out of that.
It's funny because I think that if the most alarmed people at the labs could be assured they were going to move 80% of their compute into safety and alignment as opposed to 80% into capability expansion, they would feel much better. And it sounds stopping them from doing that. And it sounds to me one thing you're actually saying is one should think of safety and alignment as capability expansion. Sure. An unsafe technology is not an advancing technology.
It's like us saying, oh, chip design is chip is chips, is RD. Chip verification is not RD. We spend most of our cost, most of our compute on verification. Emulation, verification, testing, reliability testing, lifetime testing, all of that is part of engineering. The incentives are there. The incentives are there. They are going to put their company in harm's way if they release products that harm other companies and other people.
Do you think we need liability laws that are specific to AI?