California legislators are set to vote this week on a bill that would broadly regulate the development and use of artificial intelligence in California, despite widespread opposition from many technology giants.
Here’s the background to the bill, known as SB 1047, and why it has faced strong backlash from Silicon Valley technologists and some lawmakers:
What does this bill do?
The proposal, proposed by Democrat state Senator Scott Wiener, would mandate security testing for many of the most advanced AI models, those that cost more than $100 million to develop or require a certain amount of computing power. Developers of AI software operating in the state would also have to outline ways to shut down AI models if they go wrong, effectively a kill switch.
The bill would also give state attorneys general the power to sue if developers don’t follow the rules, particularly in the case of a threat, such as AI taking over government systems like the power grid.
Additionally, the bill would require developers to hire third-party auditors to assess their security practices and provide additional protections to whistleblowers who speak out against AI abuse.
What did the MPs say?
SB 1047 has already passed the state Senate by a vote of 32-1. Last week it passed the state Assembly Appropriations Committee, setting up a vote by the full Assembly. If it passes by the end of the legislative session on August 31, it will be sent to Governor Gavin Newsom by September 30 for signature or veto.
Weiner, who represents San Francisco, home to OpenAI and many of the startups developing the powerful software, said legislation is needed to protect the public before advances in AI become uncontrolled or unregulated.
However, a group of California congressional Democrats are opposing the bill, including Nancy Pelosi of San Francisco, Ro Khanna, whose congressional district spans much of Silicon Valley, and Zoe Lofgren of San Jose.
Pelosi this week called SB 1047 “ill-informed” and said it could do more harm than good. In an open letter last week, Democrats said the bill could lock out developers from the state and threaten so-called open-source AI models, which rely on code that is freely available for anyone to use or modify.
What do tech leaders say?
Tech companies developing AI — which can respond to prompts with fully formed text, images or audio as well as complete repetitive tasks with minimal intervention — have called for stronger protections for AI deployments. They have cited the risk that the software could one day evade human intervention and lead to cyberattacks, among other concerns. But they also largely objected to SB 1047.
Wiener amended the bill to appease tech companies, partly with input from AI startup Anthropic — which is backed by Amazon and Alphabet. Among other changes, he eliminated the formation of a government AI oversight committee.
Wiener also eliminated criminal penalties for perjury, though civil suits can still be brought.
Alphabet’s Google and Meta expressed concerns in letters to Weiner. Meta said the bill threatens to make the state hostile to AI development and deployment. Yann LeCun, the Facebook parent’s chief scientist, called the bill potentially harmful to research efforts in a July 2017 post.
OpenAI, whose ChatGPT is credited with intensifying the frenzy over AI since its wide release in late 2022, has said AI should be regulated by the federal government and that SB 1047 creates an uncertain legal environment.
Of particular concern is the scope of this bill to apply to open-source AI models. Many technologists believe that open-source models are crucial to building low-risk AI applications more quickly, but Meta and others have raised concerns that they could be held responsible for overseeing open-source models if the bill passes. Wiener has said that he supports open-source models and one of the recent amendments to the bill raised the standard for which open-source models fall under its provisions.
The bill also has support in the technology sector. Geoffrey Hinton, widely known as the “Godfather of AI”, former OpenAI employee Daniel Kokotzalou and researcher Yoshua Bengio have said they support the bill.
(Except for the headline, this story has not been edited by NDTV staff and is published from a syndicated feed.)
