\n<\/p>\n
California State Senator Scott Wiener on Wednesday introduced new amendments<\/a> to his latest bill, SB 53, that would require the world\u2019s largest AI companies to publish safety and security protocols<\/a> and issue reports when safety incidents occur.<\/p>\n If signed into law, California would be the first state to impose meaningful transparency requirements onto leading AI developers, likely including OpenAI, Google, Anthropic, and xAI.<\/p>\n Senator Wiener\u2019s previous AI bill, SB 1047,<\/a> included similar requirements for AI model developers to publish safety reports. However, Silicon Valley fought ferociously against that bill, and it was ultimately vetoed by Governor Gavin Newsom<\/a>. California\u2019s Governor then called for a group of AI leaders \u2014 including the leading Stanford researcher and co-founder of World Labs, Fei Fei Li \u2014 to form a policy group and set goals for the state\u2019s AI safety efforts.<\/p>\n California\u2019s AI policy group recently published their final recommendations<\/a>, citing a need for \u201crequirements on industry to publish information about their systems\u201d in order to establish a \u201crobust and transparent evidence environment.\u201d Senator Wiener\u2019s office said in a press release that SB 53\u2019s amendments were heavily influenced by this report.<\/p>\n \u201cThe bill continues to be a work in progress, and I look forward to working with all stakeholders in the coming weeks to refine this proposal into the most scientific and fair law it can be,\u201d Senator Wiener said in the release.<\/p>\n SB 53 aims to strike a balance that Governor Newsom claimed SB 1047 failed to achieve \u2014 ideally, creating meaningful transparency requirements for the largest AI developers without thwarting the rapid growth of California\u2019s AI industry.<\/p>\n \u201cThese are concerns that my organization and others have been talking about for a while,\u201d said Nathan Calvin, VP of State Affairs for the nonprofit AI safety group, Encode, in an interview with TechCrunch. \u201cHaving companies explain to the public and government what measures they\u2019re taking to address these risks feels like a bare minimum, reasonable step to take.\u201d<\/p>\n The bill also creates whistleblower protections for employees of AI labs who believe their company\u2019s technology poses a \u201ccritical risk\u201d to society \u2014 defined in the bill as contributing to the death or injury of more than 100 people, or more than $1 billion in damage. <\/p>\n Additionally, the bill aims to create CalCompute, a public cloud computing cluster to support startups and researchers developing large-scale AI.<\/p>\n Unlike SB 1047, Senator Wiener\u2019s new bill does not make AI model developers liable for the harms of their AI models. SB 53 was also designed not to pose a burden on startups and researchers that fine tune AI models from leading AI developers, or use open source models.<\/p>\n With the new amendments, SB 53 is now headed to the California State Assembly Committee on Privacy and Consumer Protection for approval. Should it pass there, the bill will also need to pass through several other legislative bodies before reaching Governor Newsom\u2019s desk.<\/p>\n On the other side of the U.S., New York Governor Kathy Hochul is now considering a similar AI safety bill,<\/a> the RAISE Act, which would also require large AI developers to publish safety and security reports.<\/p>\n The fate of state AI laws like the RAISE Act and SB 53 were briefly in jeopardy as federal lawmakers considered a 10-year AI moratorium on state AI regulation<\/a> \u2014 an attempt to limit a \u201cpatchwork\u201d of AI laws that companies would have to navigate. However, that proposal failed in a 99-1 Senate<\/a> vote earlier in July.<\/p>\n \u201cEnsuring AI is developed safely should not be controversial \u2014 it should be foundational,\u201d said Geoff Ralston, the former president of Y Combinator, in a statement to TechCrunch. \u201cCongress should be leading, demanding transparency and accountability from the companies building frontier models. But with no serious federal action in sight, states must step up. California\u2019s SB 53 is a thoughtful, well-structured example of state leadership.\u201d<\/p>\n Up to this point, lawmakers have failed to get AI companies on board with state-mandated transparency requirements. Anthropic has broadly endorsed the need for increased transparency into AI companies<\/a>, and even expressed modest optimism about the recommendations<\/a> from California\u2019s AI policy group. But companies such as OpenAI, Google, and Meta have been more resistant to these efforts.<\/p>\n Leading AI model developers typically publish safety reports for their AI models, but they\u2019ve been less consistent in recent months. Google, for example, decided not to publish a safety report for its most advanced AI model ever released,<\/a> Gemini 2.5 Pro, until months after it was made available. OpenAI also decided not to publish a safety report for its GPT-4.1 model<\/a>. Later, a third-party study came out that suggested it may be less aligned than previous AI models<\/a>.<\/p>\n SB 53 represents a toned-down version of previous AI safety bills, but it still could force AI companies to publish more information than they do today. For now, they\u2019ll be watching closely as Senator Wiener once again tests those boundaries.<\/p>\n<\/div>\n