California, home to some of the world’s leading artificial intelligence (AI) technology companies, is often a battleground between tech lobbyists and AI regulators. So far, Congress has not enacted any comprehensive legislation to regulate the development and use of AI. Thus, proponents of AI regulations are looking towards California’s jurisdiction, aiming to enact legislation that would govern Silicon Valley. In September 2024, Governor Gavin Newsom signed AI bills to combat deepfakes and protect actors, while also vetoing a landmark AI safety bill.
The presidential election has raised concerns with regard to deepfakes spreading misinformation through deceptive content online. The signing of AB 2655 would require large online platforms to label or delete deceptive content and authorize candidates and officials to seek injunction relief against online platforms.
Earlier this year, AI-generated sexually explicit pictures of celebrities began to spread on the internet. Twitter/X, for instance, had to temporarily suspend “Taylor Swift” searches after explicit AI images of the singer circulated on the platform. SB 926 aims to target this deepfake content, making it a crime to create and distribute sexually explicit AI-generated images of a real person.
The governor also signed two bills to protect the digital likeliness of performers. The growing concerns over the use of AI were a driving force behind last year’s actors’ strike, and these bills respond to those same concerns. The first bill, AB 1836, would prohibit any commercial use of deceased actors’ voices or likeness without the consent from their estates. The second bill, AB 2602, requires contracts to specify the use of AI-generated digital replicas of the performer’s likeness or voice.
But perhaps the biggest news as of this September is Newsom’s veto of the strictest AI regulation bill proposed to date. This marked a huge win for tech companies such as Google, Meta, and OpenAI. The proposal, SB 1047, is a sweeping AI regulation bill that requires safety testing of AI models and allows prosecutors to take actions against companies whose AI models cause serious harm. It also required developers to implement a kill switch in the event that their AI becomes unreasonably dangerous.
The bill was authored by Sen. Scott Wiener, with support from Star Wars director J.J Adams, prominent AI researcher and the “Godfather of AI” Geoffrey Hinton, and owner of xAI Elon Musk. The proponents argued that the bill would merely formalize the commitments that tech companies had already made voluntarily. Lobbyists from large tech companies and prominent politicians, including Rep. Nancy Pelosi, heavily opposed this bill, stating that it would impede innovation and competition. They also argued that prosecutors should pursue those who use the technology for harm rather than developers of the technology itself.
In his rejection, Governor Newsom stated that SB 1047 focused solely on the most expensive and large-scale models and failed to consider the environment in which the AI system is deployed or its use of sensitive data and critical decision-making. While Governor Newsom vetoed the current bill, he is in favor of AI safety protocols and promised to deliver a follow up legislation with a more evidence-based approach. Without regulations from Congress, California has the stage to set the national standard for AI in the future.
