A bill to regulate artificial intelligence developed in California is close to passing. SB-1047the Safe and Secure Innovation for Frontier Artificial Intelligence Models Act has been passed by the Legislature with a 41-9 vote on Wednesday, as reported by Politico.
After a procedural vote, the go to Governor Gavin Newsom’s office to become law or be blocked by the end of September.
The bill is an attempt to curb the growing power of generative AI (AI tools that can generate information and answers by combing through the data they are trained on) being developed by companies like OpenAI, Google, Apple, Meta and many others.
The so-called Doomer bill has been approved by Elon Musk But politicians including Nancy Pelosi, tech companies and venture capitalists oppose it, saying the bill will stifle innovation in the birthplace of many of these technologies.
The bill requires safety testing for AI models that cost more than $100 million to develop or require significant computing power.
AI companies would also have to build in a safety net to prevent AI from causing unrest. They would also be overseen by the state attorney general, who would have the authority to litigate over compliance.
The bill also requires AI companies to hire outside auditors and provide whistleblower protections.
“SB-1047 would stifle AI development in California, harm business growth and jobs, and break with the state’s long tradition of promoting open-source innovation. This bill is well-intentioned, but not yet ready to become law,” a Meta spokesperson said in an emailed statement.
Apple, Google and OpenAI did not immediately respond to a request for comment.
Since it was introduced in February by Democratic Senator Scott Wiener, who represents San Francisco, the bill has been drawn discussion and concerns about how it could impact the state’s tech industry. A coalition of several tech-focused groups, including Chamber of Progress, NetChoice and Silicon Valley Leadership Group, sent an open letter to Newsom urging him to veto the bill.
