Anthropic's AI-safety pacing plan draws support — and skepticism

A public debate over whether frontier AI companies are serious about slowing down has split industry leaders after Anthropic CEO Dario Amodei published a plan to “pace the frontier” of AI development — even as Nvidia CEO Jensen Huang has publicly echoed President Donald Trump’s claims that the AI backlash is a hoax and regulation is unnecessary. On the latest episode of TechCrunch’s Equity podcast, hosts Kirsten Korosec, Sean O’Kane, and Anthony Ha debated whether executives like Amodei and OpenAI CEO Sam Altman actually mean it.

Both Ha and O’Kane were struck by how quickly other industry leaders, including Altman and, to some degree, Elon Musk, coalesced around Amodei’s framing — a shift, in O’Kane’s case, from arguing a week earlier that a slowdown wasn’t coming because of how these companies are structured. Ha noted the underlying proposals had reportedly been circulating in the AI safety community for a while rather than originating solely with Amodei, which he called encouraging as a sign of consensus, but also a reason for skepticism that it represents genuine steps toward slowing down. Both agreed the plans are still short on specifics about the risks involved and about what “slowing down” would actually mean in practice.

Korosec laid out what the general proposals actually involve: independent third-party evaluators working inside companies like Anthropic and OpenAI to monitor AI safety practices and incidents, and a second proposal for major AI companies in democratic countries to coordinate on safety standards, limits, and international cooperation. She noted consensus can be a good thing, but warned it could also mark the start of something resembling a cartel among frontier labs.

The pushback has been pointed, particularly from Huang, whose comments and an on-stage call from Trump at the All-In Summit drew scrutiny from the panel. O’Kane argued Huang has positioned himself as the industry’s “adult in the room,” a role he said shifted away from Microsoft CEO Satya Nadella after an earlier Nadella pledge to pressure Google competitively never played out — arguably the opposite occurred. Korosec pushed back that Huang’s staged-feeling appearance underscored how much Nvidia benefits from AI development continuing unpaced, and Ha noted that Trump’s and Huang’s interests are aligned, with Huang telling Trump what he wants to hear. Ha also pointed out that Altman and Amodei consistently use the word “pace” rather than “slowdown.”

Pressed on whether existing regulation and market competition alone provide enough of a safeguard, O’Kane was skeptical: “We have a federal government at the moment that certainly does not seem eager to enforce regulations broadly speaking, let alone specifically speaking here.” He added that there isn’t much consumer choice driving the market, since companies increasingly earn revenue from enterprise customers unlikely to switch providers over principle, cushioned further by heavy investor funding that lets them absorb losses.

Why it matters

The panel’s core disagreement — whether a wave of voluntary safety pledges reflects real caution or a coordinated public-relations response — is central to whether any of the proposed AI-safety mechanisms end up with teeth.