Anthropic is pressing U.S. states to adopt stricter artificial intelligence regulations in 2026, arguing that transparency laws passed just last year are no longer adequate to address the risks posed by the most powerful AI systems.
The company backed the first wave of frontier AI safety legislation in the United States in 2025, helping secure transparency and self-reporting requirements in California and New York that much of Silicon Valley opposed. Now, Anthropic says those measures have already been outpaced by rapidly advancing AI capabilities.
“The transparency-focused safety bills of 2025 were a really important start, but as the capabilities of AI systems continue to advance quickly — the policy responses need to match,” said Cesar Fernandez, Anthropic’s head of U.S. state and local government relations. “We think that transparency and self reporting are no longer sufficient safety measures for the most powerful AI systems.”
Anthropic has since endorsed an Illinois measure requiring AI labs to undergo third-party safety audits, and most recently backed a Massachusetts bill that would impose similar auditing requirements and allow the state’s attorney general to seek injunctive relief against non-compliant companies.
The push for tougher regulation comes from a company now valued at nearly $1 trillion. Anthropic’s founding mission is “to ensure that the world safely makes the transition through transformative AI,” and its leaders have framed aggressive regulation as consistent with that goal — including rules designed to prevent AI from contributing to financial disasters or mass deaths.
Not everyone accepts that framing. David Sacks, a technology adviser to President Donald Trump, has accused Anthropic of pursuing regulatory capture — using safety arguments to saddle smaller rivals with burdensome rules. Fernandez disputes this, noting that the bills Anthropic supports apply only to companies that have spent hundreds of millions on AI development and generate more than $500 million in annual revenue. “It’s hard to imagine a startup meeting that threshold,” he said.
However, several well-funded AI companies — including Safe Superintelligence, Thinking Machines Lab, and Mistral — have each raised billions from investors and could potentially approach those thresholds, raising questions about how narrow that carve-out truly is. The debate is playing out as Congress has stalled on federal AI legislation, leaving states to set the pace.
Source: WIRED