Wiener and Rutinel Win House Seats to Build America's First Federal AI Safety Law
California and Colorado's AI Safety Law Authors Enter Congress With a Federal Agenda

Two of America's most consequential state AI safety lawmakers are heading to Washington. Senator Scott Wiener of California and Representative Manny Rutinel of Colorado won their House races on Tuesday, bringing with them firsthand experience drafting and passing AI safety legislation and an explicit plan to introduce a federal version in 2027.
Their arrivals mark a major shift in the federal AI policy landscape. For years, Congress has struggled to pass any meaningful AI governance legislation, leaving states to fill the void. Now, two of the architects of those state-level experiments are walking into the chamber with existing bills in hand.
Legislative History
Wiener authored California's SB 1047, the Safe and Secure Innovation for Frontier AI Models Act, which passed both chambers of the California legislature with strong bipartisan margins in 2024 before Governor Gavin Newsom vetoed it in September of that year. Newsom stated the bill did not take into account "whether an AI system is deployed in high-risk environments, involves critical decision-making or the use of sensitive data."
Wiener did not stop. He returned the following session and authored California's SB 53, the Transparency in Frontier Artificial Intelligence Act, which Newsom signed on September 29, 2025. The law, which took effect on January 1, 2026, requires frontier AI developers to publicly disclose safety frameworks, report major safety incidents to state authorities, and establish whistleblower protections for employees who flag risks. It applies to AI models trained using more than 10²⁶ floating-point operations.
Rutinel authored Colorado's SB 24-205, the Colorado Artificial Intelligence Act, which Governor Jared Polis signed in May 2024 as the first comprehensive state AI law in the country — though Polis signed it with stated reservations and urged the legislature to revisit it. The law faced mounting legal pressure, including a federal lawsuit filed by Elon Musk's xAI on April 9, 2026, seeking to block enforcement on First Amendment and due process grounds. With xAI's challenge in court and the original law's June 30, 2026 effective date approaching, the Colorado legislature replaced it with SB 26-189, which Governor Polis signed on May 14, 2026. The replacement law scaled back Colorado's framework to focus on transparency and notice requirements for automated decision-making technologies, taking effect January 1, 2027.
Federal Plans
Both lawmakers have been explicit about their federal ambitions. Wiener told Bloomberg in a post-election statement that he intends to introduce a federal AI safety bill in 2027. Rutinel campaigned in Colorado's competitive 8th Congressional District on a platform that included federal AI governance as a top priority.
Their approach would likely draw from both states' experiences — the incident-reporting and transparency requirements of California's SB 53 and the harm-prevention framework originally embedded in Colorado's first AI act. Neither lawmaker has released formal legislative text for their federal proposal.
The federal context they are entering is a complicated one. The Trump administration has pursued AI deregulation as a policy priority. In December 2025, President Trump signed Executive Order 14365, which directed federal agencies to promote AI development with a light regulatory touch. That executive order formed the basis for the Trump National AI Legislative Framework, a four-page non-binding set of legislative recommendations released by the White House on March 20, 2026, which urged Congress to avoid regulations that could impede AI development.
Veto Arithmetic
A federal AI safety law faces long odds even with Wiener and Rutinel in Congress. Democrats appear on track to control approximately 255 House seats — a majority, but not the 290 votes required to override a presidential veto. The Senate, where Democrats appear set to hold a 55–44 majority following the 2026 midterms, would still need to overcome a potential presidential veto.
That math means any federal AI safety bill would likely require either bipartisan support or a future change in administration to become law. Wiener and Rutinel have both indicated awareness of this constraint, suggesting they intend to use their legislative positions to build a record and shift the terms of the debate, even if enactment remains uncertain in the near term.
The challenge of passing federal AI legislation is not new. Senate efforts have stalled repeatedly. In September 2026, Senator Mark Warner introduced the Artificial Intelligence Risk Management and Security Act of 2026, co-sponsored by Senators Brian Schatz and Andy Kim — only to have Senator Ted Cruz block it on the Senate floor on September 29, 2026. Earlier, in June 2026, Representatives Jay Obernolte and Lori Trahan released a 269-page "Great American AI Act" discussion draft — a comprehensive proposal that was not formally introduced as legislation before the end of that Congress. And in July 2026, Representatives Ted Lieu and Nathaniel Moran introduced the AI Kill Switch Act, a bipartisan bill proposing emergency shutdown mechanisms for advanced AI systems posing catastrophic risks.
Election Context
The victories by Wiener and Rutinel were part of a broader 2026 midterm wave that saw AI safety emerge as an organized electoral priority. According to Bloomberg's reporting, Public First, a group backed in part by Anthropic CEO Dario Amodei who contributed $1 million personally, spent more than $63 million backing 17 candidates with favorable views on AI safety across the 2026 cycle; 10 of those candidates won their races.
Total spending tied to AI safety as an electoral issue exceeded $110 million in the 2026 midterm cycle, according to campaign finance tracking cited by Bloomberg. Wiener's campaign raised approximately $3.5 million, and Rutinel's raised approximately $2.6 million.
The cycle also featured a notable personnel development in the AI safety space. Jacob Coxon, who had served as a senior AI policy official at Anthropic, resigned from the company on September 8, 2026 — a departure that drew attention in AI policy circles given its timing ahead of the midterms.
The OpenAI government affairs team had separately signaled its own interest in federal AI legislation. In a September 9, 2026 post, OpenAI policy chief Chris Lehane urged Congress to pass AI safety legislation, framing the post-midterm period as a window of opportunity for federal action. The company spent more than $2.22 million on federal lobbying in the first six months of 2026.
Congress passed the TAKE IT DOWN Act, signed by President Trump on May 19, 2025, which criminalized the nonconsensual sharing of intimate imagery including AI-generated deepfakes and required platforms to remove such content within 48 hours — the most significant federal AI-adjacent legislation to clear both chambers to that point.
Failed Federal Bills
The lack of a comprehensive federal AI safety framework has been a consistent feature of congressional sessions since at least 2023. While individual AI-adjacent bills have passed — including the TAKE IT DOWN Act — broad safety legislation has repeatedly stalled.
Senate Democrats and some Republicans have proposed multiple frameworks. The Warner-Schatz-Kim bill was the most recent casualty, blocked by Cruz before it could come to a vote. The Obernolte-Trahan discussion draft generated significant attention but never advanced to a floor vote.
Opponents of federal AI safety legislation, including Cruz, have argued that regulation risks hampering American AI competitiveness at a moment when the United States is competing with China for AI dominance. That framing has shaped much of the Senate debate and is likely to remain central to any future legislative fight.
The Senate voted 99-1 on July 1, 2025 to strip a proposed 10-year moratorium on state AI regulation from President Trump's "One Big Beautiful Bill Act" — a signal that even Republican senators were unwilling to fully preempt state-level AI governance at a time when federal alternatives did not exist.
Wiener and Rutinel enter Congress as the two sitting lawmakers with the deepest personal track records on AI safety legislation. Whether that translates into legislative success in a narrowly divided Congress under a deregulatory administration remains to be seen.



