Big Tech Opposes California’s AI Regulation Bill Despite Call for Stronger Guardrails
California is on the brink of enacting one of the country’s most comprehensive artificial intelligence (AI) regulation bills, but tech giants are pushing back against the proposed legislation. The bill, SB 1047, advanced by State Senator Scott Wiener, aims to set stringent standards for AI development and deployment, prompting significant debate among lawmakers, tech leaders, and industry insiders.
SB 1047 would impose broad regulations on advanced AI models, specifically targeting those that are costly to develop—over $100 million—or require substantial computing power. Key provisions of the bill include mandatory safety testing, the implementation of a kill switch for malfunctioning AI systems, and the appointment of third-party auditors to evaluate safety practices. The bill also empowers the state attorney general to take legal action against non-compliant developers, particularly if AI systems pose an ongoing threat, such as interference with critical infrastructure.
The legislation has already passed the state Senate with a resounding 32-1 vote and cleared the Assembly appropriations committee. It is poised for a vote by the full Assembly before the end of August, with a potential decision by Governor Gavin Newsom expected by the end of September.
Senator Wiener, representing San Francisco—a hub for AI innovation—argues that SB 1047 is crucial for safeguarding public interests as AI technologies become increasingly advanced. “We need to put protections in place before AI systems become too unwieldy or uncontrollable,” Wiener asserted.
However, the bill faces strong opposition from several California Congressional Democrats, including Nancy Pelosi, Ro Khanna, and Zoe Lofgren. Critics argue that SB 1047 may drive AI developers out of the state and hinder the growth of open-source AI models. Pelosi described the bill as “ill-informed,” while the opposing Democrats expressed concerns that it could undermine the state’s competitive edge in AI research and development.
In response to industry concerns, Wiener has made concessions, including removing plans for a government AI oversight committee and eliminating criminal penalties for perjury. Despite these adjustments, major tech companies remain critical of the bill. Alphabet’s Google and Meta have expressed worries that the legislation could make California less attractive for AI development. Meta’s chief scientist, Yann LeCun, labeled the bill as potentially detrimental to research efforts.
OpenAI, known for its ChatGPT model, has called for federal regulation instead of state-specific laws, arguing that SB 1047 introduces legal uncertainty. The bill’s impact on open-source AI models—vital for fostering innovation—has also raised concerns among technologists. Meta and others fear being held responsible for regulating open-source models, though recent amendments have increased the threshold for what qualifies as covered under the bill.
Despite the opposition, the bill has garnered support from notable figures in the tech community, including Geoffrey Hinton, Daniel Kokotajlo, and Yoshua Bengio, who see it as a step toward responsible AI governance.
As California prepares to make a decision, the debate over SB 1047 highlights the complex balance between fostering innovation and ensuring safety in the rapidly evolving field of artificial intelligence.