United States - New York - Deepfakes Regulation (Chapter 56/2024)

Overview

The New York FY2025 Budget, officially enacted as Chapter 56 of the Laws of 2024, includes a significant provision addressing the burgeoning concern of artificial intelligence-generated deceptive content, commonly known as deepfakes, in political discourse. Specifically, Part RR of this comprehensive budget bill introduced a new Section 14-107-a to the New York Election Law. This legislative amendment mandates stringent disclosure requirements for political communications that utilize synthetic media, aiming to safeguard the integrity of the electoral process and protect voters from misinformation. The law targets content that falsely depicts candidates or party officials engaging in actions or making statements they did not, especially within the critical 60-day period preceding a primary or general election. The enactment of this provision reflects a proactive stance by New York State to adapt its legal framework to the rapid advancements in AI technology and its potential for misuse in shaping public opinion and influencing election outcomes.

The impetus for this legislative action stemmed from a growing awareness of AI's potential to manipulate voter perceptions through deceptive practices, including incidents where AI-generated audio clips have been used to spread misinformation. By requiring clear and conspicuous disclaimers on such synthetic media, New York seeks to empower voters with the necessary information to discern authentic content from AI-generated fabrications. This measure is a crucial step in promoting transparency in political campaigning and holding individuals accountable for the responsible use of advanced technological tools. The law also carves out specific exemptions for content clearly identified as satire, parody, or bona fide news reporting, acknowledging the importance of free speech while mitigating the risks of intentional deception. The New York State Board of Elections is designated as the primary enforcement authority, tasked with assessing civil penalties for violations.

Definitions

Central to the application of New York Election Law § 14-107-a are several key definitions that delineate the scope and intent of the deepfakes provisions. "Synthetic media" is precisely defined as an image, audio, or video recording of an individual that has been created or significantly altered by artificial intelligence, machine learning, or similar advanced technologies, in such a way that it appears to be an authentic recording of that individual. This definition is crucial for distinguishing between genuinely manipulated content and other forms of media alteration. The law explicitly targets instances where such synthetic media depicts a candidate or a party official engaging in actions or statements they did not actually make, particularly when distributed with the intent to injure their reputation or electoral prospects, or to deceive voters.

Furthermore, the term "political communication" is broadly interpreted to encompass any communication that advocates for or against a candidate for public office, a political party, or a ballot measure. It also includes communications that merely refer to a candidate, political party, or ballot measure, provided they are distributed within sixty days of a primary or general election. This temporal and thematic scope ensures that the law primarily addresses deepfakes with direct relevance to active electoral campaigns. The requirement for a "prominent disclaimer" is also a critical definitional element, specifying that the communication must clearly state: "This [image/video/audio] has been manipulated or generated by artificial intelligence." The law's precision in these definitions is intended to provide clear guidance to political actors and the public, minimizing ambiguity in its application and enforcement.

Governance and Institutional Framework

The governance and institutional framework for the New York deepfakes provisions is primarily anchored by the New York State Board of Elections (NYSBOE). This agency is explicitly tasked with the enforcement of Election Law § 14-107-a, granting it the authority to investigate alleged violations and assess civil penalties. The NYSBOE plays a critical role in interpreting the nuances of the law and developing practical guidance for its implementation. This includes issuing advisories to political campaigns, candidates, and other stakeholders on how to comply with the disclaimer requirements and avoid engaging in deceptive practices involving synthetic media. The Board's existing mandate to oversee election integrity and campaign finance provides a natural fit for this new responsibility, leveraging its established infrastructure and expertise in regulating political communications.

Beyond direct enforcement, the NYSBOE is also expected to engage in public education efforts to raise awareness among voters about the prevalence and potential impact of AI-generated misinformation. This proactive approach is vital for fostering an informed electorate capable of critically evaluating political content, especially during heightened election periods. While the law designates the NYSBOE as the primary enforcement body for civil penalties, the broader landscape of AI regulation in New York also involves the Attorney General's office, which has issued its own guidance on identifying and reporting AI-generated election misinformation. This multi-faceted approach, involving both specific electoral oversight and broader consumer protection/public awareness initiatives, underscores New York's commitment to addressing the complex challenges posed by deepfakes in the democratic process.

Key Focus Areas

The New York deepfakes provisions concentrate on several critical areas to mitigate the risks associated with synthetic media in political contexts. A primary focus is on transparency and disclosure, mandating that any political communication containing synthetic media must carry a prominent disclaimer. This disclaimer, stating "This [image/video/audio] has been manipulated or generated by artificial intelligence," is designed to immediately alert the audience to the altered nature of the content. The law specifically targets communications that falsely depict a candidate or party official, aiming to prevent the spread of fabricated narratives that could sway public opinion or damage reputations during an election cycle. The requirement for prominence ensures that the disclaimer is not easily overlooked, thereby maximizing its effectiveness in informing voters.

Another key focus area is the scope of application, both in terms of content and timing. The law applies to political communications distributed within sixty days of a primary or general election, recognizing this period as particularly sensitive for voter decision-making. This temporal limitation ensures that the regulation is narrowly tailored to address the most immediate electoral threats. Furthermore, the law includes explicit exemptions for satire, parody, and bona fide news reporting, provided they are clearly identified as such. These exemptions are crucial for protecting legitimate forms of expression and journalistic endeavors, balancing the need for electoral integrity with fundamental free speech principles. The emphasis on intent – specifically, the intent to injure a candidate's reputation or electoral prospects, or to deceive voters – is also a critical element, distinguishing malicious deepfakes from other forms of digital content creation.

Implementation Framework

The implementation framework for New York's deepfakes provisions relies heavily on the established mechanisms of the New York State Board of Elections (NYSBOE). Upon the enactment of Part RR of Chapter 56 of the Laws of 2024, the provisions became effective immediately, indicating a swift transition to active enforcement. The NYSBOE is responsible for developing and disseminating guidance to ensure compliance across all levels of political campaigning, from statewide races to local contests. This guidance typically includes detailed explanations of what constitutes "synthetic media" and "political communication" under the law, examples of compliant disclaimers, and procedures for reporting alleged violations. The Board's existing infrastructure for handling campaign finance complaints and election law violations provides a ready-made system for processing deepfake-related complaints, ensuring that the new law can be put into practice without significant delays in establishing new bureaucratic structures.

Furthermore, the implementation framework anticipates a degree of self-regulation within political campaigns and media organizations. By clearly outlining the requirements and penalties, the law incentivizes proactive measures by political actors to audit their communications for synthetic media and ensure proper disclaimers are in place. The NYSBOE's role extends to receiving and investigating complaints from the public, candidates, or political parties regarding non-compliant communications. This complaint-driven enforcement mechanism allows the Board to focus its resources on actual or potential violations that could impact election integrity. The Board also has the authority to promulgate further rules and regulations as needed to clarify and enhance the enforcement of the deepfakes provisions, demonstrating a flexible framework capable of adapting to evolving technological challenges and practical implementation experiences.

Monitoring and Evaluation

Monitoring and evaluation of the New York deepfakes provisions, while not explicitly detailed with a formal review schedule within the enacted law, are implicitly part of the ongoing responsibilities of the New York State Board of Elections (NYSBOE). The Board, as the primary enforcement authority, will continuously monitor the landscape of political communications for the emergence of synthetic media and assess the effectiveness of the mandated disclaimers. This involves tracking the number and nature of complaints received regarding deepfakes, the outcomes of investigations, and the penalties assessed. Such data collection will be crucial for understanding the prevalence of deepfake misuse in New York elections and whether the current legal framework is adequately deterring deceptive practices. The NYSBOE's regular reporting on election activities and compliance issues will likely incorporate observations and statistics related to deepfake incidents, providing an informal mechanism for evaluating the law's impact.

Beyond internal monitoring, the effectiveness of the deepfakes provisions will also be subject to broader public and academic scrutiny. Election integrity advocates, media watchdogs, and legal scholars will undoubtedly analyze the law's application, its impact on political discourse, and its success in preventing voter deception. This external evaluation can provide valuable feedback to the NYSBOE and the state legislature, potentially leading to future amendments or refinements of the law. For instance, if certain types of synthetic media or specific platforms prove particularly challenging to regulate under the current framework, or if the penalties are deemed insufficient, legislative adjustments might be considered. The dynamic nature of AI technology necessitates an adaptive approach to regulation, meaning that continuous monitoring and periodic informal or formal evaluations will be essential to ensure the law remains relevant and effective in achieving its objectives of transparency and electoral integrity.

Penalties, Liability, and Appeals

The New York deepfakes provisions establish clear penalties for non-compliance, primarily focusing on civil liabilities rather than criminal charges for political deepfakes. Any person who violates the provisions of Election Law § 14-107-a is subject to a civil penalty. For a first violation, the penalty is capped at one thousand dollars ($1,000). Subsequent violations incur a more substantial penalty, not to exceed five thousand dollars ($5,000). These civil penalties are to be assessed by the New York State Board of Elections (NYSBOE), which holds the authority to investigate alleged breaches of the law and determine appropriate sanctions. The intent behind these penalties is to deter the creation and dissemination of deceptive synthetic media in political communications, ensuring that the financial consequences are significant enough to encourage adherence to the disclaimer requirements without being overly punitive for initial infractions.

Regarding liability, the law places responsibility on the "person" who distributes, disseminates, or publishes, or causes to be distributed, disseminated, or published, any political communication containing non-compliant synthetic media. This broad language can encompass individuals, political campaigns, committees, and potentially media entities, depending on their role in the distribution chain and their knowledge of the content's synthetic nature and deceptive intent. While the law does not explicitly detail an appeals process within Election Law § 14-107-a, decisions made by the NYSBOE are generally subject to administrative review and potential judicial challenge under New York's administrative procedure laws. This ensures that individuals or entities facing penalties have avenues to contest findings or the severity of assessed fines, upholding principles of due process. The Attorney General's office also plays a role in broader AI election misinformation, offering hotlines for reporting and potentially taking action against severe deceptive practices.

Relationship to Other Instruments

The New York deepfakes provisions exist within a broader legal and regulatory landscape, interacting with several other instruments at both the state and federal levels. Primarily, it amends the New York Election Law, specifically adding a new section (14-107-a) to Article 14, which governs campaign receipts and expenditures and other aspects of election integrity. This integration means that the deepfakes provisions are now an integral part of the state's comprehensive framework for regulating political campaigns, drawing upon existing enforcement mechanisms and definitions within election law. The law's focus on "political communication" and its temporal scope (60 days before an election) align it closely with established campaign regulations designed to ensure fairness and transparency during critical electoral periods.

Furthermore, these provisions complement other legislative efforts in New York to address AI-generated content. For instance, New York has also enacted laws related to non-consensual sexual deepfakes, establishing a right of action against such content and making the dissemination of certain intimate deepfakes a crime. While the Election Law amendment specifically targets political deception, these other laws demonstrate a broader state-level commitment to regulating harmful AI applications. At the federal level, there is no comprehensive federal law specifically addressing deepfakes in political campaigns, making state-level initiatives like New York's crucial. While federal laws like the Communications Decency Act (CDA) Section 230 offer some liability protections for online platforms, New York's law focuses on the creators and disseminators of the content, and the platforms' responsibilities are generally addressed through separate guidance or future legislation. The law also navigates the complex interplay with First Amendment rights, carefully balancing the need to prevent voter deception with protections for free speech, particularly through its exemptions for satire, parody, and news reporting.

International Alignment

As a state-level regulation within the United States, the New York deepfakes provisions do not directly align with specific international treaties or frameworks. However, the concerns it addresses – the potential for AI-generated synthetic media to undermine democratic processes and spread misinformation – are globally recognized and are the subject of increasing discussion and regulatory development in numerous countries and international bodies. Many nations are grappling with how to balance technological innovation with the need to protect public discourse and individual rights from AI misuse, particularly in electoral contexts. New York's approach, emphasizing transparency through disclaimers and establishing civil penalties, reflects a common strategy being explored by jurisdictions worldwide.

While not a direct alignment, New York's law contributes to a growing global trend of jurisdictions developing legal responses to deepfakes. For example, the European Union's Artificial Intelligence Act, while broader in scope, also includes provisions related to transparency for AI-generated content. Other countries are similarly considering or have enacted laws requiring disclosures for synthetic media, especially in political advertising. The New York law's focus on election integrity and voter protection aligns with the broader international objective of safeguarding democratic institutions from foreign and domestic interference facilitated by advanced AI technologies. Its implementation and effectiveness will likely be observed by other jurisdictions as they continue to refine their own regulatory strategies for AI and deepfakes.

Implementation Timeline

Compliance Checklist

Sources and References