one year on
California Governor Newsom signs SB 53, requiring frontier AI companies to publish safety frameworks
The first-in-the-nation law mandates transparency, incident reporting, and whistleblower protections for developers like OpenAI, Anthropic, Meta, and Google DeepMind.
Originally published Published on this site
SACRAMENTO — Governor Gavin Newsom today signed SB 53, the Transparency in Frontier Artificial Intelligence Act, making California the first US state to require large frontier AI developers to publish safety frameworks, report critical safety incidents, and protect whistleblowers.
The law applies to companies like OpenAI, Anthropic, Meta, and Google DeepMind. They must publicly describe how they incorporate national and international safety standards, and report critical safety incidents — such as autonomous cyberattacks or deceptive model behavior — to state authorities. A new CalCompute consortium will develop a framework for a public computing cluster.
The bill is Senator Scott Wiener’s second attempt at AI regulation after Newsom vetoed the broader SB 1047 last year amid industry pushback. This time, Wiener engaged major labs early, and while Anthropic endorsed the bill, OpenAI published an open letter urging Newsom not to sign. Tech firms have warned that a patchwork of state-level rules could stifle innovation.
The state’s dominant AI sector — home to 32 of the 50 top AI companies in the world — now faces its first binding transparency mandate. For an industry that spent the past year warning against a state-by-state patchwork, the question tonight is whether California just wrote the template the rest of the country ends up following.
The record
Said the law strikes a balance between protecting communities and allowing the AI industry to thrive.
Expressed gratitude for the Governor's leadership and partnership in refining the legislation.
Wrote an open letter discouraging the Governor from signing SB 53.
Endorsed the bill.
Supported the law as moving toward 'trust but verify' principles.
One year later — open only if you can handle spoilers
SB 53's core transparency and whistleblower duties took effect on January 1, 2026, and California — not Washington — became the de facto lead US regulator of frontier AI. The fight quickly moved to federal preemption: OpenAI, Meta, Google and a16z pressed for a national framework to override state-by-state rules, even as other states such as Colorado and Texas advanced their own. Through mid-2026 it stood as the first enacted frontier-AI transparency regime in the country — a template others cited and much of the industry lobbied to supersede.
Filed
Compiled and edited by Adam Allcock.
The Weekly Replay · free by email
This week, one year ago — every Sunday.
One email each Sunday: the week's replayed AI news, with the one-year-later annotations included. Written like it's breaking — dated like it isn't.
Free · double opt-in · unsubscribe anytime · privacy