What changed
California state senator Scott Wiener, who recently won a congressional race in San Francisco, has expressed deep concerns about the rapid development of advanced AI and the need for immediate federal regulation. He cited a recent report from the AI nonprofit METR, which suggested that Anthropic may have achieved recursive self-improvement—a phenomenon where AI contributes to its own advancement, potentially leading to uncontrollable progress. Wiener stated that the report highlights the urgency of regulatory action, as he believes Congress is not moving quickly enough to address these risks.
What this means in practice
Wiener has already influenced AI policy at the state level through legislation such as SB 1047 and SB 53, both of which have inspired federal proposals. He advocates for a strong, empowered federal regulator, whistleblower protections, testing and auditing requirements, and a kill switch mechanism. He also supports state-level AI regulation, arguing that states should maintain the ability to act independently of federal preemption. Wiener’s campaign received significant support from Public First Action, a nonprofit backed by Anthropic, which has led to criticism from opponents who claim he is too close to the AI industry. Wiener disputes this, stating that his support from Public First is due to his pioneering role in AI regulation.
Limitations and availability
Wiener acknowledges the challenges of passing meaningful AI legislation in Congress, citing its poor track record on tech regulation. He also expresses concerns about President Donald Trump’s stance on AI risks, which he views as a potential obstacle. Despite these challenges, Wiener remains cautiously optimistic about Congress’s ability to act, though he acknowledges the risks of inaction. The supplied announcement does not specify the exact timeline for potential regulatory measures or the current status of proposed federal legislation.
