Anthropic dropped his support for the first wave of border AI security regulations last year in the United States, providing recent transparency requirements in California and Up-to-date York that much of Silicon Valley fought against, arguing they would stifle the AI boom. However, Anthropic says these regulations may already be old-fashioned, and the company is now urging countries to adopt even more stringent regulations.
“The transparency-focused 2025 security bills were a really important start, but as the capabilities of AI systems rapidly increase, policy responses need to be aligned,” Cesar Fernandez, head of state and local government relations at Anthropic, told WIRED in an interview. “We believe that transparency and self-reporting are no longer sufficient security measures for the most powerful AI systems.”
Being so pro-regulatory is an odd message coming from a startup that is now valued at nearly $1 trillion. But Anthropic is a strange company. As we have already written, Anthropic’s leaders believe that in order to fulfill its assumptions, it must build a huge business based on the development and sale of access to advanced artificial intelligence mission: “to ensure the world safely transitions through transformative AI.”
As Anthropic has grown, it has come to support some of the nation’s toughest proposed regulations on pioneering artificial intelligence companies. Many of these policies are aimed at mitigating the catastrophic risks associated with AI, including the possibility that advanced models could contribute to financial disasters or mass deaths.
In addition to the self-reporting provisions in California AND New YorkAnthropic also supported Illinois’ measure requiring AI labs to submit their security processes to independent auditors. Recently, the company approved, among others: Massachusetts Politics it would also require a third-party audit of artificial intelligence labs and would authorize the state’s attorney general to seek injunctions from companies that don’t follow the rules.
I met with Fernandez this week to understand where the company’s AI policy is heading. Fernandez joined Anthropic earlier this year, having previously served as head of U.S. state government relations at sports betting giant FanDuel and as a senior public policy specialist at Uber. He helped both companies win political battles in states across the country. His expertise will likely prove valuable to Anthropic as Congress continues to delay passing AI legislation and states take the lead.
“Dubious” motives
While Anthropic’s pro-regulation agenda has been praised by AI safety groups, labor unions and other company allies, some Silicon Valley leaders are interpreting its political efforts through a more nefarious lens.
David Sacks, a former White House AI czar and current technology adviser to President Donald Trump, said Anthropic is essentially seeking to pass burdensome regulations that would drag smaller AI startups into bureaucracy, thus securing its leadership position in the AI race.
“Anthropic is employing a sophisticated regulatory capture strategy based on fear mongering,” Sacks wrote in: post on social media last year. “He is primarily responsible for the state’s regulatory madness that is hurting the startup ecosystem.”
Fernandez vehemently denies the accusations, noting that Anthropic only supports state AI bills that apply to “large AI modellers” – a term that is defined differently in each bill but generally refers to companies that have spent hundreds of millions of dollars developing AI and have annual revenues of more than $500 million. “It’s hard to imagine a startup meeting this threshold,” Fernandez says.
However, in a world where billions of dollars are now required to develop a competitive, pioneering AI model, there are likely a handful of promising startups that could soon reach these thresholds. To name just a few, Secure superintelligenceThinking Machines Lab and Mistral have raised billions of dollars from investors, although their revenues are still significantly lower than those of companies such as Anthropic and OpenAI. Of course, these are not ordinary startups, but they are potential competitors of Anthropic.