California’s governor vetoed proposed AI security bill

California Governor Gavin Newsom vetoed the controversial bill SB 1047 regulating artificial intelligence on September 29. The bill “does not provide a flexible, comprehensive solution to mitigate potential catastrophic risks,” the governor’s office wrote. The announcement included alternative measures to both support California’s AI industry and prevent harm.

Newsom: Bill ‘may give public false sense of security’

SB 1047, the Secure Frontier Artificial Intelligence Model Innovations Act, would be the strongest regulation in the country for generative AI. It aimed to provide protection for industry whistleblowers, order large AI developers to allow their models to be completely disabled, and hold major AI companies accountable for adhering to rigorous security protocols.

The bill passed the California State Assembly and Senate in August.

In his statement vetoing the bill, Newsom said SB 1047 “establishes a regulatory framework that may give the public a false sense of security in controlling this rapidly evolving technology” because the bill focuses on large, expensive models as opposed to smaller, high-performance models. -risky situations.

“Despite good intentions, SB 1047 does not address whether Al is deployed in high-risk environments, involves critical decision-making, or uses sensitive data,” Newsom wrote. “Instead, the bill imposes rigorous standards on even the most basic functions – provided it is implemented in a large system. I don’t think this is the best approach to protecting society from the real threats this technology poses.

However, on September 29, the voivode drew attention to several new initiatives related to generative artificial intelligence:

• The California Office of Emergency Services will need to expand its existing work in assessing potential threats related to generative artificial intelligence
• The state will convene a group of artificial intelligence experts and scientists, including Stanford University professor and artificial intelligence “grandmother” Fei-Fei Li, to “help California develop workable guardrails.”
• The state will host a meeting of academics, labor market and private sector stakeholders to “explore approaches to using GenAI technology in the workplace.”

Does California’s Artificial Intelligence Act Go Too Far or Too Little?

The primary author of SB 1047 is Senator Scott Wiener (D-Calif.). In a Sunday post on X, he criticized Newsom’s decision.

“This veto is a defeat for anyone who believes in oversight of the massive corporations that make critical decisions that impact the safety and well-being of society and the future of the planet,” he wrote.

“The governor’s veto message contains a number of criticisms of SB 1047: the bill does not go far enough, and yet it goes too far; that the risks are urgent, but we must remain cautious,” Wiener wrote in a formal response to Newsom’s decision. “SB 1047 was created by some of the top AI minds in the world, and any suggestion that it is not based on empirical evidence is patently absurd.”

The federal government has been closely monitoring California because the state could potentially set a precedent for artificial intelligence regulation. So far, it has largely refrained from implementing broad or specific regulations on artificial intelligence, opting instead for voluntary agreements.

SEE: The US government has signed an international treaty stating that artificial intelligence should respect human dignity and be subject to supervision.

Companies including OpenAI, Meta and Google opposed SB 1047 for slowing innovation or using “technically unfeasible requirements.” Other tech players, including Elon Musk and Anthropic, who helped draft the bill, supported the way the bill addresses the potential dangers of artificial intelligence.

What does the veto mean for entrepreneurs?

For business stakeholders involved in AI strategy, the veto means that large-scale AI projects in California will be subject to less state control than they otherwise would be. Newsom has approved a number of other AI-related regulations, including a ban on deepfakes during the election season and regulations on the use of AI in industries such as health care and insurance.

The veto also means that large AI models can continue to be developed in California without “kill switches.” If necessary, organizations can establish their own AI governance. In August, Deloitte found that the most important ethical issue in the implementation and development of artificial intelligence among surveyed organizations is “balancing innovation with regulation.”