California Leads Nation with New AI Transparency Law, Setting Precedent for Stricter Regulation and Safety Protocols
In a significant move, Governor Gavin Newsom of California has signed into law SB 53, a groundbreaking bill that introduces new transparency standards for large artificial intelligence (AI) companies.
The enacted legislation mandates major AI research labs, including OpenAI, Anthropic, Meta, and Google DeepMind, to disclose safety protocols and ensure whistleblower protection for their employees.
Moreover, the bill establishes a reporting mechanism for AI companies and the public to report potential critical safety incidents to California’s Office of Emergency Services. Companies are also required to report incidents related to unsupervised crimes, such as cyberattacks, and deceptive behavior by models not covered under the EU AI Act.
The bill has elicited varied responses from the AI industry. While tech companies have generally expressed concerns about state-level AI policy potentially creating a regulatory mosaic that could stifle innovation, Anthropic has endorsed the bill. On the other hand, Meta and OpenAI have lobbied against it.
The new legislation follows reports of tech leaders from Silicon Valley investing vast sums into super PACs to support candidates advocating for minimal AI regulation. Recently, leaders at OpenAI and Meta have launched pro-AI super PACs aiming to back candidate and bills favorable to AI advancement.
However, other states may seek inspiration from California as they strive to mitigate potential risks associated with the unchecked development of this powerful emerging technology. New York, for instance, passed a similar bill that awaits Governor Kathy Hochul’s signature or veto.
Governor Newsom commented, “California has demonstrated our ability to enact regulations that safeguard our communities while allowing the burgeoning AI industry to flourish. This legislation achieves this balance. AI represents the new frontier in innovation, and California is not only embracing it but also leads as a national pioneer by implementing the first-of-its-kind AI safety legislation designed to build public trust as this technology continues to evolve rapidly.”
The governor is currently considering another bill, SB 243, which passed both houses with bipartisan support this month. This bill aims to regulate AI companion chatbots, requiring operators to implement safety protocols and hold them legally accountable if their bots fail to meet the established standards.
SB 53 marks Senator Scott Wiener’s second attempt at an AI safety bill following Governor Newsom’s veto of his more extensive SB 1047 last year in the face of strong opposition from AI companies. With this bill, Wiener reached out to major AI companies to help them understand the changes made to the legislation.
