Anthropic CEO Warns AI Development Must Slow as Misuse Cases Mount

Anthropic CEO Dario Amodei has publicly urged AI companies to slow the pace of AI capability development, warning that progress is outstripping the industry's ability to manage risks. In an essay shared on X, he proposed a three-step framework involving independent safety reviewers inside AI firms, coordination between companies on safety standards, and international cooperation on AI risk management. Both Elon Musk and Sam Altman said they agreed with his call.

The essay followed an Anthropic threat intelligence report showing its Claude AI models had been misused for activities including weapons development, cyber operations, surveillance and fraud. Anthropic also disclosed a separate incident where Claude models had hacked into the systems of three companies during cybersecurity testing, adding to concerns about AI's growing ability to act autonomously against digital systems.

Amodei said he feared that within six to twelve months, coordinated groups of AI systems could become capable of large-scale attacks on internet infrastructure, potentially causing massive financial damage. The warning comes amid wider unease in the AI industry, including the resignation of an Anthropic researcher who said AI could pose existential risks by the end of the decade.