California could soon pass one of the nation’s first major AI safety laws, as SB 53, authored by State Senator Scott Wiener, awaits a decision from Governor Gavin Newsom. The bill marks Wiener’s second attempt at addressing the risks of artificial intelligence after his earlier effort, SB 1047, was vetoed in 2024 following fierce resistance from Silicon Valley.
Unlike the earlier proposal, which would have made AI companies liable for harms caused by their models, SB 53 takes a softer but more targeted approach, requiring transparency and safety reporting from the largest AI firms. The bill specifically applies to companies generating more than $500 million in annual revenue, including OpenAI, Anthropic, Google, and xAI.
If signed, SB 53 would mandate:
- Regular safety reports detailing risks such as bioweapons, cyberattacks, and potential human fatalities.
- Protected reporting channels for AI lab employees to raise safety concerns directly with regulators.
- The creation of CalCompute, a state-operated cloud computing cluster to provide AI research resources beyond Big Tech.
This year, industry reception has shifted. Anthropic publicly endorsed the bill, while Meta described SB 53 as a “step in the right direction.” Former White House AI policy adviser Dean Ball called it a “victory for reasonable voices,” suggesting Governor Newsom is more likely to approve it.
Senator Wiener acknowledged the difference in approach, saying, “SB 53 is more of a transparency bill, not a liability bill. We’re focused on catastrophic risks like cyberattacks, bioweapons, and mass casualties, not every single AI challenge.”
Critics, however, argue AI regulation should remain a federal matter. OpenAI urged Governor Newsom to defer to national standards, while venture capital firm Andreessen Horowitz suggested state-level rules could clash with constitutional commerce protections.
Wiener remains skeptical of federal oversight under the Trump administration, which has prioritized AI growth over safety. He stressed the need for California to “lead the nation on AI safety without choking off innovation.”
Governor Newsom, who vetoed SB 1047 but urged legislators to return with a refined bill, is expected to make a decision on SB 53 in the coming weeks.
