Anthropic's Amodei Pitches AI 'Frontier Pacing' Plan
Original: Dario Amodei and other AI leaders want to ‘Pace the Frontier’ but…how?
Why This Matters
Voluntary inter-lab AI safety coordination, if adopted, would set a precedent for self-regulation before governments act.
Anthropic CEO Dario Amodei, speaking at TechCrunch Disrupt on September 18, 2026, outlined a plan to 'pace the frontier' of AI development relying on independent safety evaluators and cross-lab coordination among democratic nations. The proposal has drawn both industry backing and sharp pushback from Nvidia's Jensen Huang.
One week after an Anthropic researcher's widely-cited doomsday warning shook the AI industry, CEO Dario Amodei went public with a structured response: a framework he calls 'pacing the frontier.' The plan centers on two pillars — independent third-party safety evaluators and coordinated agreements between AI labs operating in democratic countries. Amodei has not specified binding enforcement mechanisms, leaving open the central question of who ultimately polices compliance. The proposal has already attracted notable industry support, though Nvidia CEO Jensen Huang has pushed back publicly. TechCrunch's Equity podcast hosts Kirsten Korosec, Anthony Ha, and Sean O'Kane dissected the plan, focusing on whether competing labs can even agree on a shared definition of 'slowing down' — let alone submit to oversight from an external body. The episode also covers the boardroom shake-up at Automattic, WordPress's parent company, and several major deals from the week. The broader tension: Amodei is asking rivals to coordinate on restraint while the competitive race for AI capability shows no sign of easing.