① The state of California in the United States will vote on the AI regulation bill SB-1047 at the end of August. Once passed, it will be signed into law in California; ② Its stringent regulations are like a thorn in the side for AI companies in Silicon Valley, including OpenAI, Meta, Alphabet, and many scientists have expressed their opposition to the bill; ③ However, AI owner Musk has expressed support for it.
On August 27, Financial Association News (Editor Ma Lan)-The SB-1047 bill being reviewed in the state of California in the United States is dividing the AI industry in Silicon Valley into two worlds.
This plan, formally known as the "Frontier Artificial Intelligence Model Security Innovation Act", will set specific safety standards for all mainstream large models. Once a model violates the contents of this framework, developers will be held legally liable.
This bill has been seen as a major enemy by the AI community. Fei-Fei Li, a pioneer in the field of artificial intelligence and the first Sandy Briggs Chair Professor at Stanford University, pointed out that SB-1047 will harm the nascent ecosystem of artificial intelligence, punish developers unnecessarily, stifle the open-source community, hinder academic research, and not contribute to solving practical problems.
A number of technology giants, including OpenAI, Meta, and Alphabet, as well as hundreds of industry insiders, have expressed strong opposition to SB-1047. However, there are also some AI insiders who agree with regulation, and Musk is the latest and most well-known supporter.
On Monday night, Musk made a post on X, expressing his support and stating that this appeal is very difficult and will make some people discouraged, but California should pass the SB-1047 bill. Musk claims that for the past 20 years, he has been pushing for legislation on artificial intelligence, just as other technological fields are regulated.
Landmark Bill
The SB-1047 bill has already passed the Appropriations Committee of the U.S. Senate and will be voted on in the California Legislature at the end of August. Once passed, California Governor Newsom will sign it into law, directly impacting Silicon Valley.
And this bill is also seen as a leading force in AI regulation in the United States, which will have a profound impact on the future AI regulatory framework in the United States and may have ripple effects globally.
One of the regulations that the AI industry is most opposed to is that model developers are legally responsible for the downstream use and fine-tuning of the model. For example, if someone fine-tunes the GPT model and develops an AI product that contains malicious and illegal content, then OpenAI will be held responsible for this.
This provision is especially damaging to open source models, and AI leader Andrew Ng has even warned that SB-1047 will deal a devastating blow to the field of large-scale open source models.
As an alternative to this bill, on Monday, Microsoft and OpenAI announced their support for another AI bill in California, AB 3211, which requires tech companies to label AI-generated content to help combat misinformation.