California Becomes First State to Mandate AI Transparency With Signing of SB 53

California Gov. Gavin Newsom on Tuesday signed SB 53, landmark legislation that for the first time requires large AI companies to disclose their safety protocols and creates new protections for whistleblowers inside the industry.

The bill, passed by state lawmakers earlier this month, applies to major AI labs such as OpenAI, Anthropic, Meta, and Google DeepMind. It mandates that companies provide transparency into how they manage AI risks, while also establishing a mechanism for reporting potential “critical safety incidents” to the California Office of Emergency Services. That includes crimes committed without human oversight — such as AI-enabled cyberattacks — and incidents of deceptive behavior by a model, requirements that go beyond what the EU AI Act currently enforces.

In addition, SB 53 protects employees who raise safety concerns from retaliation, an increasingly pressing issue as some AI researchers have left firms like OpenAI amid disputes over safety and governance.

Tech Industry Pushback

The measure has divided the AI sector. Anthropic endorsed the bill, but both Meta and OpenAI lobbied aggressively against it. In fact, OpenAI went as far as publishing an open letter to Gov. Newsom warning that the law could slow innovation and stifle research. Broadly, companies argue that state-by-state laws risk creating a “patchwork” of regulation that complicates compliance for global firms.

The tension comes as Silicon Valley leaders have begun investing heavily in politics to shape the regulatory climate. Executives at both Meta and OpenAI have launched pro-AI super PACs, funneling hundreds of millions of dollars into supporting candidates who favor a light-touch approach to AI oversight.

National Ripple Effect

California’s move is likely to reverberate well beyond its borders. Other states are already considering similar measures: New York’s legislature has passed its own AI safety bill, now awaiting Gov. Kathy Hochul’s decision. If signed, it would further signal a willingness by states to step in while federal lawmakers in Washington remain gridlocked on comprehensive AI policy.

“California has proven that we can establish regulations to protect our communities while also ensuring that the growing AI industry continues to thrive,” Newsom said in a statement.

Calling AI “the new frontier in innovation,” he argued that the bill strikes a balance between safeguards and economic growth.

More Regulation on the Horizon

Newsom’s signature is not the end of the story. He is currently weighing SB 243, another measure that passed with bipartisan support, which would specifically regulate AI companion chatbots. That bill would require operators to implement safety protocols and hold them legally accountable if their systems fail to meet standards, responding to mounting concerns about mental health risks and exploitation by human-like bots.

For State Senator Scott Wiener, SB 53 marks a breakthrough after last year’s setback. Newsom vetoed Wiener’s earlier, broader SB 1047 following heavy industry pushback. Learning from that defeat, Wiener worked directly with AI companies to refine the new measure, a more targeted bill that still manages to set a national precedent.

With SB 53, California has planted a flag in the ongoing struggle to define how the U.S. will manage the risks of frontier AI. While federal lawmakers debate whether to leave oversight largely to industry or impose stricter national rules, Sacramento is asserting itself as a laboratory for AI governance.

If other states follow California’s lead, companies like OpenAI and Meta could soon face a patchwork of requirements across the country — a prospect they warn could fragment innovation. But to policymakers, California’s approach could also serve as a blueprint for bridging public trust and technological progress at a moment when concerns over AI’s rapid advancement are growing louder.