A game-changing bill in California has cleared a critical vote, aiming to set the first safety measures for large AI systems in the United States.
The proposal, which passed narrowly in the state assembly, now awaits a final vote in the state senate. If it passes, it will then head to Governor Gavin Newsom’s desk. Interestingly, Newsom has remained tight-lipped about his stance on the bill, although he has expressed concerns about overregulating AI in the past.
Supporters of this groundbreaking bill argue that it is essential to establish safety ground rules for massive AI models. These include systems needing more than $100 million in data to train. The measure is designed to prevent scenarios such as AI models being used to disrupt electricity grids or create chemical weapons, which, although hypothetical, are potential future risks with rapid advancements in AI technology.
Scott Wiener, the Democratic senator behind the proposal, claims that his legislation takes a ‘light touch’ approach. ‘Innovation and safety can go hand in hand – and California is leading the way,’ he remarked after the vote.
However, the bill has faced significant opposition from major tech companies like OpenAI, Google, and Meta, as well as venture capital firms. They argue that safety regulations should be a federal matter and that the bill unfairly targets developers rather than those who misuse AI.
Even so, the state of California, home to 35 of the world’s top 50 AI companies, is uniquely positioned to lead in this area. The state is already an early adopter of AI technologies and could soon use AI tools for tasks like easing highway congestion and improving road safety.
Elon Musk, the owner of the social media platform X and founder of xAI, gave his support to the bill, albeit reluctantly. He tweeted, ‘For over 20 years, I have been an advocate for AI regulation, just as we regulate any product/technology that is a potential risk to the public.’
Not everyone is on board, though. A group of California House members, including former House speaker Nancy Pelosi, has criticized the bill as ‘well-intentioned but ill informed.’ Todd O’Boyle, a senior tech policy director at the Chamber of Progress, even likened the bill to science fiction, saying it has ‘more in common with Blade Runner or The Terminator than the real world.’
Amidst the debate, Anthropic, an AI startup backed by major players like Amazon and Google, has thrown its support behind the bill. They believe it is crucial for preventing the misuse of powerful AI systems and argue that its benefits outweigh its costs.
Senator Wiener isn’t backing down either. He responded to critics by stating, ‘If they really think the risks are fake, then the bill should present no issue whatsoever.’
As California moves closer to potentially setting landmark AI regulations, the debate over the balance between innovation and safety rages on. Whether this bill becomes law or not, it has undeniably sparked a vital conversation about the future of AI and its potential risks.
Source: Theguardian