Anthropic Urges US States to Accelerate AI Regulation Amid Valuation of $1 Trillion
As Anthropic pushes for third-party safety audits and injunctive powers, industry figures accuse the company of regulatory capture to stifle competitors.
Anthropic is urging US state governments to adopt stricter artificial intelligence regulations, arguing that existing transparency laws in California and New York are already outdated. The company, valued at nearly $1 trillion, supports measures including third-party safety audits in Illinois and Massachusetts, and injunctive powers for state attorneys general.
Anthropic’s head of US state and local policy, Cesar Fernandez, stated that self-reporting is no longer sufficient for powerful AI systems. While some Silicon Valley figures accuse the company of regulatory capture to stifle competitors, Anthropic maintains its focus is on large model developers with significant revenue and development costs.
The company advocates for federal-level authority to block unsafe AI deployments, contrasting with recent executive actions by the Trump administration. This includes directives to suspend access to specific models for foreign nationals, a move the company reportedly opposed.
Anthropic has also sent a letter to the US government accusing Alibaba of a distillation attack to extract information from its models. CEO Dario Amodei has previously warned Congress about the dangers of open-source AI, though Fernandez notes current state proposals target model capabilities rather than construction methods.
Potential competitors mentioned include Safe Superintelligence, Thinking Machines Lab, and Mistral, which have raised billions but have lower revenue than Anthropic or OpenAI. Fernandez denies accusations that the regulations are designed to trap smaller startups, noting the high thresholds for development spending and annual revenue required to fall under the proposed frameworks.
