California governor vetoes AI bill amid innovation concerns

California Governor Gavin Newsom has vetoed the controversial SB 1047, known as the Safe and Secure Innovation for Frontier Artificial Intelligence Models Act.

The bill, aimed at regulating AI models through mandatory safety testing and implementing safety guardrails, was intended to address potential risks posed by emerging AI technologies.

However, Newsom argued that it would stifle innovation and fail to protect the public from the real threats associated with AI development.

In a statement issued on September 29, Newsom expressed concerns that the bill applied overly stringent standards to basic AI functions, affecting both large and small developers.

He noted that while the legislation targeted major AI firms such as OpenAI, Meta, and Google, it did not effectively address broader safety concerns related to AI.

Newsom emphasised the need for a science-based approach to AI regulation and committed to working with AI safety experts to create more practical safeguards.

The bill, introduced by San Francisco Democratic Senator Scott Wiener, had proposed requiring AI developers to implement a “kill switch” for their models and publish risk mitigation plans.

Developers would also face potential legal action if their models posed ongoing threats, such as AI system takeovers.

Despite these measures, the bill faced significant opposition from Silicon Valley, as well as politicians like House Speaker Nancy Pelosi, who argued it would hinder AI innovation.

Newsom, while vetoing the bill, acknowledged the importance of establishing effective AI safety protocols.

He highlighted that his administration had signed over 18 AI-related bills in the past month and stressed that regulators cannot wait for a major catastrophe before taking action.

Tech leaders remain divided on the issue.

While Elon Musk supported the bill, noting that AI regulation is essential, others warned that it could expand in scope and negatively impact smaller developers.