Controversial California AI Bill Clears Legislature, Heads to Governor for Approval!

The California State Assembly has passed the Safe and Secure Innovation for Frontier Artificial Intelligence Models Act (SB 1047), according to a report by Reuters.

The bill represents one of the first substantial attempts to regulate artificial intelligence in the United States and has been a major point of contention in Silicon Valley and beyond.

Veto SB 1047, Gov. Newsom! California’s economy can’t afford to stifle AI innovation with overly restrictive regulations.

— CharlesAI Cyber (@ChCyberSec) August 29, 2024

If enacted, the bill would require AI companies operating in California to adopt several safety measures before training sophisticated foundation models.

These measures include the ability to shut down an AI model quickly, safeguards against “unsafe post-training modifications,” and a mandatory testing process to assess whether a model or its derivatives are at risk of “causing or enabling a critical harm.”

Gov. Newsom should veto SB 1047. It’s a solution in search of a problem, based on sci-fi fantasies rather than real-world needs. Let’s focus on creating a business-friendly environment in California, not stifling innovation with unnecessary regulations.

— PastPresentAlex (@ppalexhist) August 29, 2024

Senator Scott Wiener, the bill’s main author, described SB 1047 as a “highly reasonable” piece of legislation that asks large AI labs to perform safety testing for catastrophic risks—something they have already committed to doing.

He highlighted the potential of AI to positively impact the world while emphasizing the importance of balancing that potential with safety.

Critics of SB 1047, including major AI companies like OpenAI and Anthropic, as well as politicians such as Zoe Lofgren and Nancy Pelosi, have argued that the bill is overly focused on catastrophic harms and could unfairly disadvantage smaller, open-source AI developers.

To address some of these concerns, the bill was amended to replace potential criminal penalties with civil ones, narrow the enforcement powers granted to California’s Attorney General, and adjust the requirements for joining a “Board of Frontier Models” created by the bill.

Curious to see how this bill will impact innovation in CA

— Farmer Sebby (@sebby_coder) August 28, 2024

The amended bill is now awaiting a final vote in the State Senate, which is expected to pass. After this vote, the bill will head to Governor Gavin Newsom, who, as noted by The New York Times, has until the end of September to either sign it into law or veto it.

Despite its passage in the Assembly, the bill remains controversial, with powerful interests on both sides of the debate.

For example, OpenAI and other tech giants have expressed concerns that the legislation could stifle innovation and drive AI companies away from California.

Is the margin too narrow to be veto-proof?

— Ryan Carey (@ryancareyai) August 29, 2024

On the other hand, some advocates, including whistleblowers from AI companies like OpenAI and Elon Musk, who runs AI firm xAI, have supported the bill, citing the potential harms AI could pose without proper regulation.

While acknowledging potential ambiguities in the bill, companies such as Anthropic have indicated that the benefits likely outweigh the costs.

The bill also mandates third-party audits to assess safety practices and includes protections for whistleblowers who expose unethical or unsafe AI practices.

Comment
byu/Sudden-Degree9839 from discussion
inaiwars

In the event of non-compliance or if an AI system poses an ongoing threat, such as taking over critical infrastructure, the California Attorney General has the authority to enforce the legislation.

Additionally, developers of AI systems are required to establish a “kill switch” mechanism to deactivate an AI model if it starts behaving dangerously.

As the AI safety bill nears Governor Newsom’s desk, its fate remains uncertain. Governor Newsom, who has previously called for a balanced approach to AI regulation, will play a pivotal role in deciding whether California becomes the first state to impose such regulations on the AI industry.

Comment
byu/Sudden-Degree9839 from discussion
inaiwars

His decision will be closely watched by supporters and opponents of the bill, who see it as a potential turning point for AI governance in California and across the United States.

SB 1047 has become a battleground for a broader discussion on how to manage the rapid development of AI technologies. While its supporters argue that it is a necessary step to ensure safety in the AI field, its detractors warn of potential economic and innovation setbacks.

Comment
by from discussion
inaiwars

The coming weeks will determine whether California will set a new standard for AI regulation or if the state will choose a different path in addressing the complex challenges posed by artificial intelligence.

For more news and trends, visit AI News on our website.