California Combat Deepfakes Act (AB 2655)

This Act, known as the Defending Democracy from Deepfake Deception Act of 2024, was enacted to address the growing threat of AI-generated disinformation, specifically 'deepfakes,' in California elections. Its primary goal is to protect the integrity of democratic processes by preventing the spread of misleading digital content that could influence voters or suppress participation.

The law focuses narrowly on election-related deepfakes and places obligations on 'large online platforms' – those with at least one million California users. It mandates that these platforms take specific actions, such as blocking or labeling deceptive content, during critical periods surrounding an election. The Act also establishes mechanisms for accountability, allowing various parties to seek legal remedies against non-compliant platforms.

The Act defines 'large online platform' as any online platform with at least one million California users in the 12 months preceding an election period. This threshold ensures the law targets entities with substantial reach and potential impact on the electorate, who also typically have the resources to implement compliance measures.

'Materially deceptive content' is defined as audio or visual media, digitally created or modified (including deepfakes and chatbot outputs), that would falsely appear to a reasonable person to be an authentic record of the depicted content. Crucially, minor modifications that don't significantly alter the perceived meaning are excluded. The Act also provides specific exemptions for content that constitutes satire or parody, as well as for legitimate broadcasting stations and regularly published online news periodicals.

Large online platforms are mandated to implement robust measures to manage election-related deepfakes. Specifically, they must block the posting or sending of 'materially deceptive content' during designated periods before and after an election in California. This is a proactive blocking requirement aimed at preventing the most harmful forms of deepfake deception from reaching voters.

For other types of content that may not warrant outright blocking but are still deemed inauthentic, fake, or false, platforms are required to apply appropriate labels. This dual approach allows for a nuanced response, ensuring voters are informed about the authenticity of content without necessarily removing all potentially misleading information, provided it is clearly identified as non-authentic. Platforms must develop and deploy advanced detection technologies and internal procedures to meet these obligations.

A critical component of the Act is the requirement for large online platforms to establish clear and accessible procedures for California residents to report content. These mechanisms allow individuals to flag content they believe has not been blocked or labeled in compliance with the Act's provisions.

This provision empowers individual citizens to participate in the oversight process, effectively acting as a distributed network of monitors. By providing a formal channel for citizen complaints, the Act aims to create a more responsive and accountable system for content moderation, ensuring that platforms are held to their obligations not just by official bodies, but also by their user base.

The primary enforcement mechanism under this Act is the authorization for various parties to seek injunctive relief against a non-compliant large online platform. This means that if a platform fails to meet its obligations to block or label deceptive content, or to establish adequate reporting procedures, aggrieved parties can petition a court to order the platform to cease its non-compliant activities.

Authorized parties include candidates for elected office, elected officials, elections officials, the California Attorney General, and district or city attorneys. A significant aspect of the Act is the directive that actions seeking such injunctive relief shall be assigned precedence in court. This prioritization is crucial for addressing deepfake deception quickly, given the time-sensitive nature of election cycles. The Act primarily establishes a civil enforcement framework and does not impose direct criminal penalties or specific monetary fines.

AB 2655 does not operate in isolation; it integrates with and amends existing California legal frameworks. Specifically, it modifies Section 35 of the Code of Civil Procedure to ensure that legal actions brought under this Act receive expedited judicial attention, aligning with the urgency typically afforded to election-related matters. It also adds Chapter 7 to Division 20 of the Elections Code, establishing the new deepfake regulations.

This Act builds upon and strengthens prior California law concerning deceptive audio or visual media in elections, which previously prohibited the distribution of such content with malice and intent to deceive voters. AB 2655 expands this by introducing proactive obligations for large online platforms to block or label content, shifting some of the responsibility from individual distributors to the platforms themselves. The exemptions for satire, parody, and legitimate news outlets, as detailed in Section 2, are also critical for balancing regulation with free speech.

The Defending Democracy from Deepfake Deception Act of 2024 was approved by the Governor and chaptered by the Secretary of State on September 17, 2024. However, the Act itself does not become legally binding and enforceable until its operative date of January 1, 2025.

This timeline provides a clear deadline for large online platforms to implement all necessary changes to their policies, procedures, and technical infrastructure to comply with the Act's requirements. The period between approval and the operative date is intended for platforms to prepare for the new regulatory landscape, ensuring they are fully compliant when the law takes effect.