California Governor Gavin Newsom has vetoed a controversial AI safety bill. The Democrat justified his decision by stating that the bill only considered the largest and most expensive AI models, without taking into account their use in high-risk situations. Newsom emphasized that smaller models could also make critical decisions, while larger models are often used for low-risk activities.
The rejected bill would have required developers of large AI models to take “reasonable care” to prevent “unreasonable risks” or “critical harm”. It would have applied to AI models that exceed certain computing power thresholds and cost at least $100 million to train. Leading tech companies such as Google, Meta, and Microsoft had raised concerns, while prominent AI scientists supported the bill.
Newsom announced that he would work with leading AI researchers on new legislation he could support. He has already signed other AI-related bills, including those concerning the labeling of AI-generated content and the regulation of election-related deepfakes. The governor emphasized the need to protect the public from real threats without unnecessarily hindering the innovative power of the technology.
Sources: Wall Street Journal, Governor Gavin Newsom