TechAnthropic CEO says it’s time to pump the brakes on AI

Anthropic CEO Dario Amodei proposed a three-step plan to slow frontier AI development, giving companies time to build safeguards and regulators time to assess models. Anthropic will grant broad model access to third-party evaluators such as METR, while Amodei also called for industry-wide and eventually global safety standards. He cited risks from recursive self-improvement and rogue AI agents.

AI developmentAI safetyAnthropic

TThe Verge★★★★☆2026-09-12 16:23Original