Gavin Newsom dropped a sweeping AI regulatory blueprint in Washington on Wednesday, framing California as the national template while he quietly builds a presidential runway.
The proposal, unveiled at a closed-door roundtable blocks from the White House, would grant state officials unprecedented authority to intervene inside private AI labs, embed independent safety monitors, and pursue “emergency brakes” capable of shutting down models deemed dangerous. Newsom called it an “action document” and said he is already working with legislators in Sacramento to turn the framework into law.
“This is an action document. I’m already working with members of our Legislature to introduce the legislative frameworks to move quickly and make real what we’re promoting and promising in this report,” Newsom told reporters at the event.
The rollout marks a sharp populist pivot for a governor who once courted Silicon Valley as an economic engine. In recent months he has signed data-center restrictions, ordered a study on “kill switch” requirements, and floated an AI sovereign wealth fund — each step widening the distance between his current posture and his 2024 veto of a major safety bill authored by then-Sen. Scott Wiener.
The Wiener veto shadow
That veto looms over the new plan. Newsom rejected Wiener’s legislation last year, citing unintended consequences for innovation, yet Wednesday’s framework embraces several concepts the earlier bill championed — including third-party audits and state intervention powers. The governor has said he does not regret the veto, but the policy convergence is hard to miss.

“This takes things to the next level, not only of conversation, but of consciousness in terms of how we effectively regulate artificial intelligence, not only in my home state, but we believe a template for how we should be doing it federally as well,” Newsom said.
A panel of insiders
The report, titled “AI Safety in 2026 and Beyond,” was fast-tracked by an executive order in September. Newsom appointed a four-member expert panel — Stanford professor Rob Reich, former Obama White House Chief Digital Officer Jason Goldman, economist Gillian Hadfield, and Alondra Nelson, who led the Office of Science and Technology Policy under President Joe Biden — and gave them a Monday deadline to deliver recommendations.
Hadfield and Nelson appeared alongside Newsom at Wednesday’s roundtable, joined by Dee Myers, head of the governor’s business and economic development office, and former Office of Management and Budget Director Shalanda Young.
Washington as the stage
Choosing Washington over Sacramento or San Francisco for the rollout was deliberate. The event put Newsom in front of a swarm of national political and tech reporters, letting him cast California as a counterweight to President Donald Trump’s resistance to strict AI regulation.
“This is not a Truth Social or a tweet,” Newsom said, framing the effort as necessary “in the absence of any federal leadership.”
What the plan actually demands
The framework calls for expanding California’s existing frontier AI safety and transparency law, SB 53, and would authorize the state to cap the use of automation in developing advanced AI systems. It also seeks to embed independent monitors directly inside labs — a provision that industry groups have historically resisted as operationally intrusive.
Legislative text is still being drafted. Newsom’s team says they want bills moving quickly when the session reconvenes, but the governor’s term ends in January 2027, leaving a narrow window to enact a regime this ambitious.
The question now is whether Sacramento will move faster than the labs.