Zuckerberg Rejects Coordinated AI Slowdown, Cites Meta’s Muse Safety Delay
Meta CEO Mark Zuckerberg pushed back Tuesday against calls for a coordinated slowdown in artificial intelligence development. He said individual AI labs have both responsibility and incentives to manage safety risks themselves. Zuckerberg cited competition, user trust and legal liability as reasons companies should prioritize safe AI systems.
His comments followed Anthropic CEO Dario Amodei’s call for companies to pace frontier AI development. OpenAI CEO Sam Altman and Elon Musk supported parts of Amodei’s proposal. Zuckerberg instead argued that labs can slow or delay products when safety requires it, without waiting for an industry-wide agreement. He pointed to Meta’s decision to delay its Muse AI agent for several months to improve safety and security.
Zuckerberg also said Meta’s AI division uses independent evaluators and described broader external evaluation as an industry best practice. The debate highlights a widening split over whether AI safety should rely primarily on company-level safeguards or coordinated action.
Why it matters
The dispute could shape how investors assess AI development costs, product timelines and regulatory risk. It also signals continued disagreement among major AI companies over safety governance.
