Anthropic CEO Says It's Time to Pump the Brakes on AI

**Anthropic CEO Calls for Slowing Down AI Development to Prioritize Safety** Dario Amodei, CEO of leading AI research company Anthropic, has publicly advocated for a significant slowdown in the pace of artificial intelligence development. In a comprehensive essay, Amodei outlined a three-step plan, which he terms "pacing the frontier," aimed at giving both developers and regulators crucial time to establish robust safeguards and thoroughly evaluate advanced AI models. Amodei's proposal emphasizes the urgent need to prioritize safety and responsible development over rapid advancement. The first step of this plan involves Anthropic unilaterally granting wide-ranging access to its models for third-party evaluators, such as METR. This move is intended to ensure the company's strict adherence to safety practices and commitments, setting a precedent for transparency and external oversight within the industry. Looking beyond Anthropic's immediate actions, Amodei envisions the second step as a broader, industry-wide collaboration. This phase would likely involve major AI developers coming together, potentially with significant government involvement, to collectively agree on and implement shared standards for slowing down development, building in safety measures, and facilitating regulatory scrutiny. While the full details of this multi-stage plan are yet to be unveiled, Amodei's call marks a pivotal moment in the ongoing debate about the future trajectory and responsible governance of artificial intelligence.