Anthropic Advocates for Stricter AI Regulations, Citing Rapid Advancement

Anthropic, a prominent player in the advanced artificial intelligence sector, is now advocating for a significant escalation in AI safety regulations across the United States. While the company previously supported the initial wave of transparency-focused laws enacted in states like California and New York last year, it contends that these measures are rapidly becoming insufficient. Cesar Fernandez, Anthropic’s head of US state and local government relations, stated in an interview that the pace of AI capability advancement necessitates a more robust policy response. "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," Fernandez explained. "We think that transparency and self reporting are no longer sufficient safety measures for the most powerful AI systems."

This assertive stance on regulation from a company reportedly valued at nearly $1 trillion might seem counterintuitive to some in the tech industry. However, Anthropic's core mission, as previously reported, is to ensure a safe transition into an era of transformative AI, which it believes requires building a substantial business around developing and providing access to advanced AI. As the company has scaled, it has increasingly lent its support to some of the most rigorous proposed regulations for frontier AI development. These proposed rules aim to address catastrophic risks, including potential AI-induced financial crises or widespread harm.

Evolving Regulatory Landscape

Beyond the self-reporting mandates in California and New York, Anthropic has also backed an Illinois initiative requiring AI labs to undergo third-party safety process evaluations. More recently, the company endorsed a policy in Massachusetts that mandates similar third-party auditing for AI labs and grants the state's attorney general the authority to seek injunctions against non-compliant companies. Fernandez, who joined Anthropic earlier this year after significant government relations roles at Uber and FanDuel, is spearheading these policy efforts at a time when federal AI regulation remains stalled, leaving states to take the lead.

Scrutiny and Accusations of Regulatory Capture

While AI safety advocates and labor unions have lauded Anthropic's regulatory agenda, some figures within Silicon Valley view its actions with suspicion. David Sacks, a former White House AI czar and technology advisor to President Donald Trump, has publicly accused Anthropic of employing a "sophisticated regulatory capture strategy based on fear-mongering." Sacks argues that Anthropic aims to burden smaller AI startups with cumbersome regulations, thereby solidifying its own leadership position. "Anthropic is running a sophisticated regulatory capture strategy based on fear-mongering," Sacks wrote on social media. "It is principally responsible for the state regulatory frenzy that is damaging the startup ecosystem."

Fernandez vehemently denies these allegations, emphasizing that Anthropic's supported legislation specifically targets "large AI model developers." This term, he notes, is defined differently in each bill but generally refers to entities that have invested hundreds of millions in AI development and generate over $500 million in annual revenue. "It’s hard to imagine a startup meeting that threshold," Fernandez stated. However, with the immense capital required to develop cutting-edge AI models, a handful of promising startups like Safe Superintelligence, Thinking Machines Lab, and Mistral, despite lower current revenues, could soon approach these thresholds, posing potential competition to Anthropic and OpenAI.

Defining the Boundaries of Regulation

Anthropic's proposed regulatory framework includes a mechanism for governments to block the deployment of new AI models deemed unsafe, a suggestion that carries a degree of irony given recent actions by the Trump administration. The administration had previously directed Anthropic to suspend access to its most powerful AI models, Mythos and Fable 5, for foreign nationals, a move Anthropic found unfavorable. The company believes that the ultimate authority to block unsafe AI model deployments should reside with the federal government, not state lawmakers, though Fernandez acknowledges this is an evolving discussion. This stance was further complicated when Anthropic wrote in a blog post that such blocks should only occur through a fair and transparent evaluation process, following the export control directive.

Furthermore, Anthropic has been active on the federal front, recently accusing Chinese tech giant Alibaba of a distillation attack aimed at extracting information from its models. Some AI researchers have countered these claims, suggesting Anthropic's true motive is to lobby the US government to ban open-weight Chinese models, thereby driving businesses towards its own enterprise solutions. Anthropic CEO Dario Amodei has previously expressed concerns to Congress regarding the risks associated with open-source AI. At the state level, Fernandez maintains that Anthropic's endorsed legislation focuses on model capabilities rather than construction, stating, "It’s less of a question of the model construction and more of the model capabilities, and when they rise to a certain level, they should be captured in these state frameworks."

Broader Societal Concerns

Regardless of the sincerity of Anthropic's motives, the company is undeniably shaping the future of AI policy. Its recent model releases have brought the cybersecurity implications of advanced AI into sharp focus. For years, Anthropic has been vocal about the catastrophic risks posed by frontier AI. However, concerns more commonly voiced by the public, such as job displacement, the environmental impact of data centers, and the effects of AI on children, have not inspired a similar legislative push from frontier AI labs. While Anthropic and its competitors have pledged to shield ordinary taxpayers from data center costs and have proposed solutions for AI-driven job displacement, the industry has dedicated less political capital to state-level laws addressing these more immediate societal issues. Fernandez indicated that Anthropic is "eager to engage with lawmakers" on these broader concerns and is already "having those conversations throughout the states," though a coordinated legislative push comparable to that for existential risks has yet to materialize.