Why California’s SB 53 might provide a meaningful check on big AI companies

The California State Senate has recently passed an important AI safety bill, known as SB 53, which is now awaiting the decision of Governor Gavin Newsom. This bill comes on the heels of a previous AI safety proposal, SB 1047, which was vetoed by Newsom last year. However, SB 53 presents a more focused approach, targeting only those AI companies that generate over $500 million in annual revenue. In a recent episode of TechCrunch’s flagship podcast, Equity, hosts Max Zeff and Kirsten Korosec discussed the potential implications of SB 53. Zeff expressed optimism that this new legislation may stand a better chance of being enacted, especially given its narrower scope and support from AI firm Anthropic. He emphasized the significance of AI safety regulations, noting that this bill could serve as a crucial check on the expanding influence of major tech companies. Zeff pointed out that SB 53 mandates AI labs to publish safety reports regarding their models and requires them to report any incidents to the government. Furthermore, it provides a safe channel for employees within these organizations to voice their concerns without fear of retaliation, despite the existence of non-disclosure agreements. Korosec highlighted the importance of California as a central hub for AI development, noting that many prominent AI firms are based in the state. While acknowledging the relevance of regulations in other states, she stressed that California's legislation could have far-reaching effects due to its concentration of AI activity. Zeff also addressed the bill's limitations, mentioning that it specifically exempts smaller startups from comprehensive regulations, which had been a point of contention in previous discussions about the potential impact on the startup ecosystem. By focusing on larger companies like OpenAI and Google DeepMind, SB 53 aims to regulate the most influential players in the AI field without stifling innovation from emerging startups. In the context of broader AI regulation, the podcast participants noted that the current federal administration has taken a more lenient stance, potentially complicating state-level efforts. This ongoing regulatory discourse may see further developments, especially as political dynamics shift in the future. This conversation reflects a growing awareness and urgency surrounding AI safety as the technology continues to evolve and permeate various aspects of society.