Anthropic urges states to adopt tougher AI regulations beyond transparency

Anthropic, the AI safety-focused startup now valued at nearly $1 trillion, is urging US states to adopt tougher regulations on frontier AI models, arguing that the transparency laws it supported in California and New York are already outdated. Cesar Fernandez, Anthropic’s head of US state and local government relations, told WIRED that self-reporting and voluntary disclosures are insufficient as AI capabilities accelerate. The company has endorsed an Illinois measure requiring third-party audits of safety processes and a Massachusetts bill that would let the attorney general seek injunctive relief against non-compliant labs.

Critics, including former White House AI czar David Sacks, accuse Anthropic of pursuing regulatory capture—pushing cumbersome rules to burden smaller rivals while securing its own market position. Anthropic denies this, noting that the bills target only developers spending hundreds of millions or generating over $500 million in revenue. However, startups like Safe Superintelligence, Thinking Machines Lab, and Mistral have raised billions and could soon meet those thresholds. Fernandez argues that any company building powerful AI should face the same rules, as the risks are universal. With Congress stalled, states are becoming the primary arenas for AI regulation.