(The Center Square) – California Gov. Gavin Newsom signaled wariness against heavy artificial regulation, telling a live audience at a major San Francisco conference that a significant piece of AI regulation heading to his desk could have a “chilling effect” on the state’s nascent AI industry.
State Sen. Scott Wiener’s SB 1047, an AI “safety” bill, passed the legislature last month and would create vast new compliance and safety regulations for large-scale AI models and fund a California government AI. Speaking at the 2024 Dreamforce conference in San Francisco with Salesforce CEO, billionaire, and campaign donor Marc Benioff, Newsom shared some of his concerns about the bill.
“We’ve been working over the last couple years to come up with some rational regulation that supports risk-taking, but not recklessness,” Newsom said. “That’s challenging now in this space, particularly with SB 1047, because of the sort of outsized impact that legislation could have, and the chilling effect, particularly in the open source community.”
The Wiener bill would require AI companies to — before training a model — implement measures to prevent “critical harm,” a term “defined as mass casualties, at least $500 million in damage, or other comparable harms.” Before making a model publicly available or using it, companies would also have to assess whether the model could cause or enable critical harm, test ways in which the imagined scenarios could occur, and take measures to prevent such events. AI companies would then need to make copies, with potential redactions, of safety and security protocols and conduct independent audits of companies’ adherence to their duties.
Upon the bill’s passage in the legislature, Wiener touted his bill as “a light touch, commonsense measure that codifies commitments that the largest AI companies have already voluntarily made.”
Technology advocates shared concerns that the new bill is too heavy-handed, and that Newsom’s words, combined with his signing of several smaller AI bills regarding deepfakes, suggest Newsom may be heading towards a veto.
“We’re definitely getting the sense that a veto might be imminent – especially considering that Governor Newsom shares our concerns about a ‘chilling effect’ on smaller developers,” said Allie Caccamo, communications manager for Chamber of Progress, a pro-technology lobbying organization, to The Center Square. “Ultimately, we’re hopeful that Newsom recognizes how SB 1047 unfairly preferences existing models at the expense of startups – further entrenching the dominance of incumbent players.”
Caccamo highlighted the differences in industry support between SB 1047 and AB 2355, a bill requiring political advertisements to disclose if AI generated or altered deepfakes are used.
“I’ll also note the contrast between SB 1047 – which forces model developers to engage in speculative fiction about imagined threats of machines run amok – and the newly-signed AB 2355,” continued Caccamo. “We’re glad to see AB 2355 enacted this week because it actually accomplishes what most AI legislation sets out to do: protecting citizens from digital harms without creating new barriers to economic growth and consumer benefits.”
Since its post-pandemic peak in September 2022, California has lost 154,000 private-sector jobs, nearly 100,000 of which were in the information sector, which includes software development, media production, internet search, data processing and hosting. San Francisco office vacancy rates are at a record 37%, leading to property value declines five times worse than the national average.
