Anthropic CEO Dario Amodei published an essay calling for AI development to proceed at a more measured pace, arguing that while continued development is not in question, the associated risks are serious enough to warrant independent monitoring and coordinated regulation. His proposal outlined a three-point plan involving third-party evaluation of AI models, industry-wide standards, and global regulation, which he said Anthropic would adopt unilaterally while urging governments to require similar commitments from other frontier labs.
OpenAI CEO Sam Altman and xAI founder Elon Musk both expressed agreement with Amodei’s position, with Altman calling independent evaluators a valuable idea and stating that current safety standards aren’t sufficient to justify pushing capabilities further. Former Anthropic researcher Jacob Coxon told the BBC that many people within AI companies privately fear the technology poses a genuine risk of human extinction within the next two years.
Amodei acknowledged that any slowdown would need to avoid sacrificing U.S. competitiveness, particularly against China, and called for restrictions on AI chip exports to authoritarian governments. Hugging Face CEO Clement Delangue announced a new initiative aimed at supporting independent AI evaluation, while some observers, including investor Chamath Palihapitiya, suggested Amodei’s proposal may also serve to consolidate Anthropic’s competitive position. Both Anthropic and OpenAI are reportedly preparing for future public offerings.
