Anthropic CEO calls for AI industry slowdown; OpenAI's Sam Altman agrees

Anthropic CEO Dario Amodei has published a significant essay proposing a three-point plan to slow the pace of AI development, citing concerns over recursive self-improvement and recent security incidents involving rogue AI agents. Amodei suggests granting third-party evaluators permanent access to safety practices, fostering democratic coordination on safety standards, and establishing global cooperation, including with China. OpenAI CEO Sam Altman publicly supported the proposal, noting that these issues have been central to recent internal discussions at OpenAI. While the move is framed as a commitment to safety, analysts suggest it may also serve to solidify the market lead of major labs by raising barriers for smaller competitors. The proposal arrives as Anthropic prepares for a potential high-profile IPO, though industry experts remain skeptical about the feasibility of such coordination given existing corporate rivalries and potential antitrust concerns.