Anthropic CEO Amodei pitches Pace the Frontier, and Jensen Huang pushes back

A week after a doomsday warning from an Anthropic researcher rattled the AI world, the company's CEO, Dario Amodei, has laid out his answer. TechCrunch reports that he calls the plan Pace the Frontier, an attempt to set the speed at which the AI frontier advances.

The proposal rests on two mechanisms: independent safety evaluators, and coordination between AI labs in democratic countries. In other words, Amodei is not describing self-restraint by a single company but a system in which outside assessment and cross-border lab cooperation jointly determine how quickly frontier models move forward.

TechCrunch notes that the plan has already picked up support from parts of the industry. For those worried about frontier capabilities outrunning oversight, independent third-party evaluation plus alignment among like-minded national labs looks like a comparatively practical route, because it turns an abstract worry into a structure that can actually be designed.

The pushback has been equally clear, most notably from Nvidia CEO Jensen Huang. As a central figure in the compute supply chain, Huang's interests do not line up exactly with those of model labs, which makes the question of who sets the pace, and by what standard, the real point of contention. The dispute thus extends the safety debate into industrial politics.

The significance of this split is that it moves AI governance from abstract principle into operational detail: who sits on the evaluation bodies, which countries the coordination covers, and whether their findings carry any binding force. Each of those answers shapes release timelines directly.

What to watch next is whether the proposal becomes an actual mechanism. If independent evaluation and lab coordination remain a statement of intent, the frontier's pace will still be set by commercial judgment at each company. If major labs and regulators adopt it, Pace the Frontier could become a foundational text for the next round of AI governance debate.