California Governor Vetoes AI Safety Law

by time news

2024-09-30 05:35:56

California Governor Gavin Newsom during a rally in Los Angeles on March 21, 2024.

Until the last minute, both stations campaigned in Silicon Valley. On Sunday evening, September 29, the governor of California, Gavin Newsom, made his decision: he vetoed a bill which intends to impose stricter protections on large models of artificial intelligence in the United States.

Read also This article is reserved for our subscribers In California, the battle over AI regulation

The issue is SB 1047, or the “law for safe innovation in pioneering AI models”, has divided the world of technologies, as well as the Californian political class. Although not as demanding as European regulations, it makes artificial intelligence companies legally liable for damages caused by their models.

He asked that they include “off the switch” (keep changing) to remove their systems if they become ineffective and cause serious damage, such as a major loss of life or property damage exceeding $500 million (€448 million). Elon Musk declared himself in favor, unlike Mark Zuckerberg (Meta), Sam Altman (OpenAI) or pro-Trump entrepreneur Marc Andreessen.

Fear of “undue impact on industry”

According to Mr. Musk, many historical figures of AI, such as Geoffrey Hinton and Yoshua Bengio, approve of the matter. “We believe that the most powerful AI models can soon pose serious threats, such as increased access to biological weapons and cyber attacks on critical infrastructure”Mr. Hinton wrote in a letter written to the governor and signed by dozens of former and current employees of the AI ​​sector.

During the annual conference of software giant Salesforce on September 24 in San Francisco, Mr. Newsom expressed his skepticism. Always quick to praise the avant-garde initiatives carried out by California, especially on climate or consumer protection, he regrets that the project on AI security replaces the federal debate on the issue.

Read also This article is reserved for our subscribers The European strategy of AI: the battle continues between content creators and technology companies

He also expressed his fear “undue impact on the industry”. “We dominate this area, and I don’t want to lose that opportunity,” he said, noting that thirty-two of the largest artificial intelligence centers in the world are located in California. When it comes to the risks posed by AI, he says he tries to stick to those that are demonstrable, as opposed to those that are rational. The proposed plans require more than $100 million to build. No model has reached this threshold, but believers believe it could be reached quickly.

#California #Governor #Vetoes #Safety #Law

You may also like

Leave a Comment