SACRAMENTO, Calif. (Diya TV) — California Gov. Gavin Newsom on Monday signed a landmark bill that sets the nation’s first safety rules for powerful artificial intelligence systems.
The Transparency in Frontier Artificial Intelligence Act, known as SB 53, requires leading AI developers to disclose how they are building safe and trustworthy models. The law also creates a system to report major AI-related risks, while protecting whistleblowers who raise public safety concerns.
Newsom called the measure a “balance between innovation and protection.” He said California must lead on AI just as it has led in technology for decades. “This legislation strikes that balance,” the governor said.
The law places new duties on top AI companies. They must publish documents that explain how they follow national and global best practices. They also must notify state officials if critical incidents occur that could pose harm. In addition, the state attorney general can enforce civil penalties for companies that fail to comply.
Sen. Scott Wiener, D-San Francisco, wrote the bill. He said the measure shows California can both back innovation and guard against risk. “With a technology as transformative as AI, we have a responsibility to support that innovation while putting in place commonsense guardrails,” Wiener said.
The law arrives as pressure grows for federal action. Last year, Newsom vetoed a separate bill from Wiener that focused more on liability. SB 53 centers instead on transparency, creating what Wiener called a “trust but verify” framework.
The new law also builds on a state report released earlier this year. Newsom had asked a group of leading academics and experts to study how California should respond to frontier AI. The report urged evidence-based rules and warned that the federal government has yet to pass a broad AI policy.
Mariano-Florentino Cuéllar, a former California Supreme Court justice, praised the law. He said it follows the report’s recommendations and keeps the United States at the forefront of technology. Stanford AI leader Fei-Fei Li and UC Berkeley dean Jennifer Tour Chayes also backed the approach.
California dominates the global AI landscape. Thirty-two of the top 50 AI companies worldwide are based in the state. Last year, 15.7% of all U.S. AI job postings came from California, far ahead of Texas and New York. The Bay Area also attracted more than half of all global venture funding for AI startups.
The law creates a state-led consortium, called CalCompute, to explore the creation of a public computing cluster. The project aims to support research and expand access to safe and ethical AI tools. State officials will also review the law every year to recommend updates as technology evolves.
Industry reaction has been mixed. Some companies and trade groups warned that state laws could lead to a patchwork of regulations. Meta recently launched a super PAC to push back against what it views as heavy-handed AI rules. However, several firms expressed cautious support.
Anthropic, one of the most influential AI companies, endorsed the measure. Co-founder Jack Clark said the law sets “meaningful transparency requirements without imposing prescriptive technical mandates.” OpenAI also said the law could align with federal efforts if implemented correctly.
Despite these endorsements, critics remain wary. The Chamber of Progress and the Consumer Technology Association both argued that the rules could slow innovation. They want Congress to take the lead instead.
Meanwhile, lawmakers in Washington are beginning to act. Sens. Josh Hawley, R-Mo., and Richard Blumenthal, D-Conn., introduced a federal bill on Monday that would require top AI developers to evaluate advanced systems and report potential dangers.
The global debate is also heating up. At the United Nations last week, President Donald Trump warned that AI could bring both immense benefits and serious risks. Ukrainian President Volodymyr Zelenskyy said the world faces the most dangerous arms race in history, with AI at its core.
