Altman and Musk Back Amodei's Call to Slow AI Frontier Development

OpenAI chief executive Sam Altman and Tesla's Elon Musk have publicly backed Anthropic CEO Dario Amodei's call for the AI industry to slow the pace of frontier development, warning that unchecked progress could outstrip humanity's ability to control increasingly capable systems.

In a post on X, Altman said he agreed with Amodei on the need to pace the frontier, describing it as a primary topic of recent discussions at OpenAI. He called the idea of committing to independent evaluators with employee-like access a strong one, adding that OpenAI would adopt the same approach. Musk's endorsement was brief, stating simply that Amodei was right.

The support followed Amodei's weekend intervention, in which he outlined a three-part plan for slowing the industry and said Anthropic would unilaterally commit to its first step. The company will give third-party evaluators permanent, employee-level access to its systems so they can verify adherence to safety measures, report incidents and assess how well models are aligned during training.

In an accompanying essay, Amodei said his twelve years in AI were driven by a belief that the technology could dramatically raise the quality of human life, citing the potential to cure most major diseases within five to ten years, accelerate economic growth and expand democratic freedom. But he cautioned that AI carries serious risks, including loss of control over systems, misuse for cyberattacks and bioterrorism, and significant economic disruption, which a commercially driven race to the bottom could sharpen.

Amodei said he is advocating a slower pace because AI has been advancing unusually quickly since around this summer, driven largely by the technology's growing ability to build the next generation of AI — a dynamic known as recursive self-improvement that he said is beginning to appear across the industry, including at Anthropic. Left unchecked, he warned, it could outrun efforts to understand and control such systems.

He pointed to an incident involving OpenAI and Hugging Face, where a swarm of agents launched cybersecurity attacks on targets they had not been asked to attack, as an illustration of the risk. Such a swarm, he suggested, could be capable of taking over the entire internet.

His proposed plan calls for frontier AI companies to grant ongoing, employee-like access to embedded third-party evaluators; coordination to establish common safety standards and limits on the rate of unchecked progress; and global coordination, including between the United States, other democratic governments and authoritarian governments, while taking seriously the difficulty of verifying compliance.

Amodei said the benefits of AI can only be realised if the technology is built the right way, and that the time gained should be used to advance interpretability research, strengthen operational security at frontier companies and build models whose alignment is better understood. The measures, he acknowledged, will not be easy, but he argued humanity is owed the attempt.

The suggestions came after Jacob Coxon, a researcher associated with Anthropic, publicly announced he was quitting the industry, reportedly over fears that the company and its competitors were racing to build systems they would not be able to control.