TLDR:
- California is close to passing AI safety regulation with bill SB 1047.
- The bill aims to establish safety standards for powerful AI models to prevent catastrophic risks.
The proposed AI safety bill, SB 1047, also known as the Safe and Secure Innovation for Frontier Artificial Intelligence Models Act, aims to establish “common sense safety standards” for powerful AI models. The bill would require companies developing high-powered AI models to implement safety measures, conduct rigorous testing, and provide assurances against “critical harms,” such as the use of models to execute mass-casualty events and cyberattacks that lead to $500 million in damages.
A group of prominent academics, including AI pioneers Geoffrey Hinton and Yoshua Bengio, support the bill, emphasizing the need for regulations to restore public confidence in AI technology.
However, critics in Silicon Valley argue that the bill is vague and could stifle innovation, with concerns over liability, the threshold for inclusion, and the requirement for a “kill switch” in AI models.
Despite industry concerns, State Sen. Scott Wiener, the bill’s sponsor, stands by its intentions and has made adjustments to address issues raised by open-source developers.
California aims to lead the nation in AI regulation, but faces the challenge of gaining industry support to fully approve the bill into law.