# California AI Laws (2024)

California's 2024 AI legislation, a broad set of laws beyond SB 1047, regulates transparency, deepfakes, and algorithmic accountability, signed by Governor Gavin Newsom in September 2024.

In 2024, California enacted a series of artificial intelligence laws that extended far beyond the widely publicized Senate Bill 1047. While SB 1047, the Safe and Secure Innovation for Frontier Artificial Intelligence Models Act, dominated national headlines and was ultimately vetoed by Governor Gavin Newsom on September 29, 2024, the state legislature passed and the governor signed numerous other AI-related measures. These laws addressed a wide range of concerns, including transparency in AI-generated content, the use of deepfakes in political campaigns, and the accountability of automated decision-making systems. Together, they positioned California as a pioneer in AI regulation, even as the state's approach remained more fragmented than the comprehensive framework proposed by SB 1047.

The legislative package reflected a growing urgency among California lawmakers to respond to the rapid proliferation of [generative-ai](https://www.wikiprompt.org/wiki/generative-ai) tools. The laws were crafted amid intense lobbying from industry groups, academic institutions, and civil society organizations. Governor Newsom, a Democrat, signed the bills in a series of announcements throughout September 2024, emphasizing the need to protect consumers and voters while fostering innovation. The resulting statutes created new obligations for companies developing and deploying [artificial-intelligence](https://www.wikiprompt.org/wiki/artificial-intelligence) systems within the state, which is home to many of the world's leading AI firms, including [openai](https://www.wikiprompt.org/wiki/openai), [anthropic](https://www.wikiprompt.org/wiki/anthropic), and [google-deepmind](https://www.wikiprompt.org/wiki/google-deepmind).

## Transparency in AI-Generated Content

One of the most significant pieces of legislation was Assembly Bill 2013, which mandated transparency for training data. The law requires developers of generative AI systems to publish documentation about the datasets used to train their models. This documentation must include information about the sources of data, whether the data includes copyrighted material, and how the data was curated and processed. The law applies to developers that provide AI services to Californians, regardless of where the company is headquartered. It took effect on January 1, 2025, giving companies a short window to comply.

AB 2013 was seen as a direct response to lawsuits filed by authors and artists against companies like [openai](https://www.wikiprompt.org/wiki/openai) and [anthropic](https://www.wikiprompt.org/wiki/anthropic), who alleged that their copyrighted works were used without permission to train [large-language-model](https://www.wikiprompt.org/wiki/large-language-model)s. The law does not prohibit the use of copyrighted material, but it forces transparency about its use. Legal experts noted that the documentation requirements could be burdensome for smaller startups, while larger firms with more resources were better positioned to comply. The law also required developers to describe the purpose of the model and the intended use cases, as well as any known limitations or risks.

## Deepfakes in Political Campaigns

California also moved aggressively to regulate the use of deepfakes in elections. Assembly Bill 2655, signed into law in September 2024, requires large online platforms to label or remove deceptive AI-generated content related to elections. The law applies to content that depicts a candidate or elected official in a way that is materially deceptive, such as a fake video showing a candidate saying something they never said. Platforms must take action within 72 hours of being notified of such content. The law builds on earlier California statutes that addressed deepfakes in elections, but it extends the requirements to a broader range of platforms and content.

AB 2655 was complemented by Assembly Bill 2839, which allows courts to issue injunctions against the distribution of deceptive AI-generated election content. The law also permits candidates to seek expedited legal remedies. Together, these laws created a robust legal framework for combating AI-driven disinformation in the 2024 election cycle and beyond. Supporters argued that the laws were necessary to protect democratic processes, while critics raised concerns about potential First Amendment violations and the difficulty of defining what constitutes a "materially deceptive" deepfake.

## Algorithmic Accountability and Automated Decision-Making

Another notable law, Senate Bill 942, addressed the use of AI in automated decision-making systems. The law requires companies that use AI to make consequential decisions - such as those affecting employment, housing, credit, or healthcare - to provide consumers with meaningful information about how the system works. This includes the logic involved in the decision, the significance of the AI's role, and the consumer's right to opt out or request human review. The law took effect on January 1, 2025, and was modeled in part on the European Union's General Data Protection Regulation.

SB 942 was praised by privacy advocates as a step toward algorithmic transparency. However, it was also criticized for being vague about what constitutes a "consequential decision" and for not requiring companies to conduct independent audits of their systems. The law's enforcement mechanisms were left largely to the California Privacy Protection Agency, which was tasked with issuing regulations to implement the statute. Some industry groups warned that the law could lead to a patchwork of compliance burdens, as companies would need to adapt their systems to meet California's requirements.

## Protections for Performers and Creators

California also enacted laws aimed at protecting actors, musicians, and other performers from unauthorized AI replicas. Assembly Bill 2602 requires contracts between performers and studios to include clear language about the use of digital replicas. The law, which took effect on January 1, 2025, prohibits the use of a performer's digital replica without explicit consent, unless the use is covered by a collective bargaining agreement. It was supported by the Screen Actors Guild - American Federation of Television and Radio Artists (SAG-AFTRA), which had recently concluded a strike that included concerns about AI.

A related law, Assembly Bill 1836, extended similar protections to deceased performers. It requires consent from the estate or heirs before a digital replica of a deceased person can be used in a commercial production. The law was seen as a response to the growing use of AI to recreate the likenesses of historical figures and deceased celebrities. Both laws were notable for their specificity, addressing the unique challenges posed by AI in the entertainment industry, which is a major economic driver in California.

## The Veto of SB 1047

Despite the passage of these laws, the most ambitious piece of AI legislation in California failed to become law. SB 1047, introduced by Senator Scott Wiener, would have required developers of large AI models to implement safety protocols, including a kill switch and third-party audits. The bill also would have created a new state agency to oversee frontier AI models. Governor Newsom vetoed the bill on September 29, 2024, arguing that it was overly broad and could stifle innovation. In his veto message, Newsom said the bill focused on the largest models and could impose "stringent standards" that were not justified by current risks.

The veto was a victory for many in the tech industry, including [openai](https://www.wikiprompt.org/wiki/openai) and [anthropic](https://www.wikiprompt.org/wiki/anthropic), which had lobbied against the bill. However, it disappointed advocates who argued that California needed strong safeguards against the potential harms of advanced AI. In his message, Newsom said he had directed state agencies to work with experts to develop alternative approaches, including risk-based frameworks for AI regulation. The veto highlighted the tension between the state's desire to lead in AI regulation and its interest in maintaining its position as a global hub for technology innovation.

## Reactions and Industry Response

The 2024 laws drew mixed reactions from the AI industry. Companies like [openai](https://www.wikiprompt.org/wiki/openai) and [anthropic](https://www.wikiprompt.org/wiki/anthropic) publicly supported some of the transparency and deepfake measures, while expressing concerns about the cumulative burden of compliance. Smaller startups, particularly those in the [generative-ai](https://www.wikiprompt.org/wiki/generative-ai) space, worried that the documentation requirements of AB 2013 would be difficult to meet. Some companies began hiring compliance officers and legal teams to navigate the new regulations. Industry groups, such as the Chamber of Progress and the TechNet trade association, argued that the laws could create a fragmented regulatory environment, as other states and the federal government were also considering AI legislation.

Civil society organizations, including the American Civil Liberties Union and the Electronic Frontier Foundation, were generally supportive of the transparency and accountability measures. They argued that the laws gave consumers and voters important tools to understand and challenge AI-driven decisions. However, some privacy advocates noted that the laws did not go far enough, particularly in requiring independent audits or prohibiting certain high-risk uses of AI altogether. The debate over AI regulation in California is likely to continue, as the state legislature has already signaled that it will consider additional AI bills in 2025.

## Implementation and Enforcement

The implementation of the 2024 laws falls largely to state agencies, including the California Privacy Protection Agency and the Attorney General's office. The Privacy Protection Agency is tasked with writing regulations for SB 942, which will clarify the scope of the law and the procedures for enforcement. The agency has held public workshops to gather input from stakeholders, including industry representatives, academics, and consumer advocates. The Attorney General's office is responsible for enforcing the deepfake laws and can bring civil actions against violators.

Compliance with the new laws is expected to be a significant undertaking for companies operating in California. The state's large technology sector, which includes major AI developers and cloud providers like [amazon-web-services](https://www.wikiprompt.org/wiki/amazon-web-services), [azure](https://www.wikiprompt.org/wiki/azure), and [google-cloud](https://www.wikiprompt.org/wiki/google-cloud), will be most affected. Many of these companies have already begun to publish model documentation and implement content-labeling systems in anticipation of the laws. The long-term impact of the laws will depend on how they are interpreted by courts and regulators, as well as on whether the federal government passes its own AI legislation that could preempt state laws.

## Broader Context and Future Outlook

California's 2024 AI laws are part of a broader global movement toward AI regulation. The European Union's AI Act, which was formally adopted in 2024, imposes comprehensive requirements on AI systems based on their risk level. Other countries, including China and the United Kingdom, have also introduced AI governance frameworks. In the United States, the federal government has taken a more piecemeal approach, with executive orders and agency guidance rather than comprehensive legislation. California's laws, while narrower than the EU's AI Act, represent a significant step in the absence of federal action.

The future of AI regulation in California remains uncertain. The veto of SB 1047 demonstrated the political challenges of passing comprehensive AI legislation, even in a state with a Democratic supermajority. However, the passage of the 2024 package showed that there is bipartisan support for addressing specific harms, such as deepfakes and algorithmic bias. As AI technology continues to evolve, California lawmakers will likely revisit the issue, potentially introducing new bills to address emerging challenges. For now, the 2024 laws provide a foundation for AI governance in the state, balancing innovation with consumer protection.

---
Source: https://www.wikiprompt.org/wiki/california-ai-laws-2024
License: CC BY-SA 4.0 (https://creativecommons.org/licenses/by-sa/4.0/)
Last updated: 2026-09-12T16:24:01.320265+00:00