California Governor Gavin Newsom is facing a crucial decision regarding a bill that aims to prevent artificial intelligence (AI) systems from causing catastrophic events. The bill, known as SB 1047, proposes holding AI vendors liable if their products are used to cause significant harm. However, Newsom expressed concerns about the bill, stating that he is interested in AI legislation that solves current issues without disrupting California’s booming AI industry.
During a conversation with Salesforce CEO Marc Benioff at the 2024 Dreamforce conference, Newsom emphasized the need for rational regulation that supports risk-taking while avoiding recklessness. He acknowledged the challenges in the AI space and highlighted the potential outsized impact and chilling effect that SB 1047 could have, particularly within the open-source community.
One of the main criticisms of SB 1047 is that it focuses on preventing AI’s role in mass casualty and cybersecurity events costing over $500 million. Critics argue that the bill fails to address the short-term issues that AI systems are currently creating and may stifle innovation. This criticism has gained traction within the tech industry, with attendees at various tech conferences expressing concerns about SB 1047.
Despite his reservations about SB 1047, Newsom has demonstrated a commitment to regulating AI. On the same day, he signed five bills into law that address AI-related problems, such as AI-generated election misinformation and the creation of AI clones of actors by Hollywood studios. These bills likely represent the “demonstrable risks” that Newsom mentioned during the conference.
Newsom understands the importance of California’s leadership in the absence of federal regulation. He highlighted the state’s previous success in regulating tech issues like social media and privacy and expressed a desire to maintain California’s dominance in the AI sector. However, he also emphasized the need to address extreme concerns surrounding AI technology responsibly.
While Newsom acknowledged that signing SB 1047 may not immediately disrupt the AI industry, he expressed concerns about the long-term impact of signing the wrong bills. He emphasized the need for careful consideration to avoid jeopardizing California’s position in the AI landscape.
As of now, Newsom has not made a definitive decision on whether he will sign or veto SB 1047. OpenAI, Nancy Pelosi, the United States Chamber of Commerce, and Big Tech trade groups are urging him to veto the bill. On the other hand, Elon Musk and Anthropic have shown tepid support, while renowned AI researchers like Yoshua Bengio and Geoffrey Hinton have fully endorsed SB 1047.
Governor Newsom has two weeks to make his decision, and the future of SB 1047 remains uncertain. However, based on his recent remarks, the prospects for the bill may not be promising. It is clear that Newsom recognizes the importance of AI regulation but is cautious about the potential consequences of SB 1047.