Governor Newsom vetoes ambitious AI safety bill

Governor Newsom Vetoes Ambitious AI Safety Bill

Sacramento, CA – On Sunday, Governor Gavin Newsom vetoed Senate Bill 1047 (S.B. 1047), an artificial intelligence safety bill aimed at addressing the regulation and control of large A.I. systems. The legislation, introduced by Senator Scott Wiener (D-San Francisco), sought to implement safety measures on advanced artificial intelligence models before their public release. Newsom’s veto has halted what would have been one of the most comprehensive AI regulatory frameworks in the United States.

The proposed bill was notable for its stringent requirements. Among its key provisions were mandates for AI developers to perform safety tests on large AI systems and submit detailed safety plans to the state attorney general. It also included stipulations granting the attorney general the authority to hold companies accountable for significant harm, such as death or property damage, resulting from their technologies. Additionally, the bill proposed the installation of “kill switches” that could deactivate AI systems in cases of severe threats, including potential biowarfare and mass casualties.

However, Governor Newsom argued that the bill’s focus was overly narrow, primarily targeting expansive, frontier AI models and failing to consider smaller, specialized systems that could also pose risks. “I do not believe this is the best approach to protecting the public from real threats posed by the technology,” Newsom stated. “The bill applies stringent standards to even the most basic functions — so long as a large system employs it.”

The decision has drawn sharp reactions from various factions. Advocates of the bill, including tech experts, academics, and several high-profile industry leaders, were fierce in their support. Notable endorsements came from SpaceX CEO Elon Musk, the Center for AI Safety, and prominent researchers in the AI field. The bill was also backed by Hollywood figures, notably actors Pedro Pascal and Alec Baldwin, and TV producer Shonda Rhimes, who collectively called for the bill to be signed into law to safeguard public wellbeing.

Conversely, opposition to the bill emerged from significant tech industry players, including Meta, the parent company of Facebook, and OpenAI, the developers behind ChatGPT. Critics argued that the bill’s measures could significantly stifle innovation. Moreover, prominent Democratic figures, including Nancy Pelosi (D-San Francisco), expressed concerns about the bill’s potential implications on technological progress.

In his statement, Governor Newsom acknowledged the difficulty in crafting effective AI regulations that balance innovation with public safety. He highlighted the fast-paced evolution of AI technology and the need for regulations that adapt to its diverse applications. “SB 1047 does not take into account whether an AI system is deployed in high-risk environments, involves critical decision-making or the use of sensitive data,” Newsom noted.

Governor Newsom has pledged to continue working on AI safety measures. His administration is engaging with leaders in the AI field to develop practical protections that address the technology’s capabilities and associated risks. This collaboration aims to devise rational regulations that support innovation while preventing reckless deployment of AI systems.

Despite the veto, SB 1047’s supporters remain hopeful that the issue will be revisited and refined in future legislative sessions. Overcoming the governor’s veto would require a two-thirds majority vote from both houses, a challenging but not impossible feat.

The debate over SB 1047 underscores the broader, ongoing conversation about artificial intelligence regulation in a rapidly changing technological landscape. As lawmakers and industry leaders grapple with this evolving field, the future of AI regulation remains a critical concern for many stakeholders.

Source: Noah Wire Services