United States - California - AI-Generated Image Crimes (SB 926)

Overview

California Senate Bill 926 (SB 926), officially titled "Crimes: distribution of intimate images," represents a pivotal legislative effort to address the burgeoning threat of artificial intelligence (AI) being used to generate non-consensual sexually explicit content. Introduced by Senator Aisha Wahab, this bill was signed into law by Governor Gavin Newsom on September 19, 2024, and is slated to take effect on January 1, 2025. The core objective of SB 926 is to amend California Penal Code Section 647, thereby expanding the scope of existing "revenge porn" laws to encompass images created or substantially altered through digitization, commonly referred to as deepfakes. This legislative update acknowledges the rapid advancements in generative AI technology, which has made it increasingly easy to produce highly realistic, yet entirely fabricated, intimate images of individuals without their consent. The bill specifically targets the intentional creation and distribution of such images when they are designed to appear authentic and are known, or should be known, to cause serious emotional distress to the person depicted.

The enactment of SB 926 underscores California's proactive stance in regulating the ethical implications and potential harms associated with AI. Prior to this legislation, existing laws primarily focused on the non-consensual distribution of actual intimate images. However, the rise of deepfake technology created a significant loophole, as fabricated images did not always fit the legal definition of authentic content, making prosecution challenging. SB 926 closes this critical gap by explicitly criminalizing the creation and distribution of digitally manipulated sexually explicit images, providing victims with essential legal tools to seek justice against perpetrators of digital harassment and exploitation. This move is particularly significant given statistics indicating that a vast majority of non-consensual deepfakes disproportionately target women, highlighting the bill's importance in protecting vulnerable populations from severe emotional and reputational harm.

Definitions

Central to the application of California SB 926 are several key definitions that delineate the scope of the newly established crime. The term "digitization" is crucial, defined as the act of altering an image in a realistic manner using an image or images of a person other than the person depicted, or through the use of computer-generated images. This definition is specifically crafted to capture the essence of AI-generated content, ensuring that images fabricated or heavily manipulated by artificial intelligence systems fall under the purview of the law. It distinguishes these digitally created images from traditional photographs, thereby addressing the technological advancements that necessitate this legislative update. The inclusion of "computer-generated images" explicitly covers content produced entirely by AI, without necessarily relying on source images of another person.

Furthermore, the bill refers to a "photorealistic image," which is broadly defined to include any photo realistic image, digital image, electronic image, computer image, computer-generated image, or other pictorial representation. The critical qualifier for such an image is that it must be "created in a manner that would cause a reasonable person to believe the image is an authentic image of the person depicted." This "reasonable person" standard is an objective test, designed to assess whether the deepfake is convincing enough to mislead an average observer into believing it is genuine. The legislation also references "intimate body part or parts" and specific sexual acts, aligning with existing definitions within Penal Code Section 647 concerning non-consensual intimate images. The element of "serious emotional distress" is another vital component, requiring that the person distributing the image knows or should know that its distribution will cause significant psychological or emotional harm to the depicted individual, and that the victim indeed suffers this distress. This subjective and objective standard ensures that the law targets malicious intent and actual harm, rather than accidental or innocuous distributions.

Governance and Institutional Framework

The governance and institutional framework for enforcing California SB 926 primarily rests within the existing legal and judicial systems of California. As an amendment to Penal Code Section 647, the bill integrates directly into the established criminal justice apparatus. This means that local law enforcement agencies, including municipal police departments and county sheriff's offices, will be responsible for investigating reports of violations. Once investigations are complete, district attorneys and city attorneys will be tasked with prosecuting offenders. The bill's classification of the offense as a misdemeanor ensures that it falls within the jurisdiction of local courts and prosecutors, who are already familiar with handling cases under Penal Code Section 647. This approach leverages existing resources and expertise, facilitating a more streamlined implementation of the new provisions without requiring the creation of entirely new regulatory bodies or enforcement mechanisms dedicated solely to AI-generated content.

The legislative intent behind embedding SB 926 within the Penal Code is to provide law enforcement with clear, actionable tools to combat digital harassment and exploitation. By building upon established legal precedents for non-consensual intimate images, the bill aims to provide a robust framework for addressing the specific challenges posed by AI-generated deepfakes. This framework relies on the judiciary's interpretation of terms such as "reasonable person" and "serious emotional distress," which are common in legal statutes and allow for flexibility in application to diverse factual scenarios. The California Attorney General's office may also play a role in providing guidance and coordination for statewide enforcement efforts, particularly in cases that may span multiple jurisdictions or involve complex digital forensics. The overall institutional design emphasizes integrating AI-related offenses into traditional criminal law, reflecting a strategy to adapt existing legal structures to new technological realities rather than creating a separate, parallel regulatory system for AI.

Key Focus Areas

California SB 926 primarily focuses on two critical aspects of AI-generated sexually explicit content: its creation and its distribution. The bill explicitly makes it a crime for a person who is 18 years of age or older to intentionally create and distribute, or cause to be distributed, any photorealistic image or other pictorial representation of an intimate body part or sexual act of another identifiable person. This dual focus is crucial because it addresses both the source of the harmful content and its spread. Criminalizing creation aims to deter individuals from producing such material in the first place, recognizing the inherent harm in its very existence, regardless of whether it is widely disseminated. Simultaneously, prohibiting distribution targets the act that directly inflicts widespread emotional distress and reputational damage upon victims. This comprehensive approach acknowledges that the harm from deepfakes can stem from both their initial generation and their subsequent circulation across digital platforms.

Another key focus area is the intent and impact of the deepfake. The law requires that the image be created in a manner that would cause a reasonable person to believe it is an authentic image of the person depicted. This objective standard ensures that the law applies to convincing deepfakes, rather than clearly identifiable parodies or artistic representations. Furthermore, a crucial element for prosecution is that the person distributing the image knows or should know that its distribution will cause serious emotional distress to the depicted individual, and that the person depicted actually suffers that distress. This establishes a clear link between the perpetrator's actions, their awareness of the potential harm, and the actual harm experienced by the victim, aligning the deepfake statute with the principles of existing revenge porn laws. The bill also includes exemptions for content related to matters of public concern or public interest, as well as distributions made in the course of lawful public proceedings or court orders, balancing individual protections with constitutional rights and legitimate public discourse.

Implementation Framework

The implementation framework for California SB 926 is designed to seamlessly integrate into the state's existing criminal justice system, leveraging established legal processes and enforcement mechanisms. Upon its effective date of January 1, 2025, the provisions of SB 926 will become an enforceable part of California Penal Code Section 647. This means that victims of non-consensual AI-generated sexually explicit images will be able to report these crimes to local law enforcement agencies, including police departments and sheriff's offices. Law enforcement will then be empowered to investigate these reports, gather evidence, and, where appropriate, make arrests. The evidence collection process may involve digital forensics to trace the origin and distribution of the deepfake content, a task that requires specialized skills and resources.

Once an investigation is complete and sufficient evidence is gathered, prosecutors—such as district attorneys or city attorneys—will be responsible for bringing charges against the alleged perpetrators. Since the offense is classified as a misdemeanor, cases will typically be heard in local courts, following standard criminal procedure. The prosecution will need to demonstrate that the defendant intentionally created and/or distributed the deepfake, that it appeared authentic to a reasonable person, and that the defendant knew or should have known it would cause serious emotional distress, which the victim subsequently suffered. This framework provides a clear pathway for victims to seek legal redress and for the state to hold offenders accountable. Furthermore, the bill's emphasis on expanding existing "revenge porn" laws ensures that legal practitioners and the judiciary have a foundational understanding upon which to interpret and apply the new provisions, facilitating a consistent and effective implementation across the state. The effectiveness of this framework will also depend on public awareness campaigns to inform potential victims of their rights and the avenues for reporting such crimes, as well as training for law enforcement and legal professionals on the technical aspects of AI-generated content.

Monitoring and Evaluation

While California SB 926 itself does not explicitly detail a formal monitoring and evaluation framework, its effectiveness will inherently be subject to ongoing scrutiny by legislative bodies, law enforcement agencies, and victim advocacy groups. The impact of the law will likely be assessed through various indirect metrics, such as the number of reported incidents, arrests, and successful prosecutions related to AI-generated sexually explicit images. Data collected by law enforcement on these cases will be crucial for understanding the prevalence of this crime and the efficacy of the legal tools provided by SB 926. Furthermore, victim support organizations will play a vital role in monitoring the law's practical application, identifying any challenges victims face in reporting crimes or navigating the justice system, and advocating for further legislative refinements if necessary. The dynamic nature of AI technology necessitates continuous observation of how deepfake capabilities evolve and whether the current legal definitions remain adequate to address emerging forms of misuse.

Legislators, particularly those on relevant committees such as Public Safety and Judiciary, will likely conduct periodic reviews to evaluate the law's impact and consider potential amendments. This iterative process is common in rapidly evolving technological domains, where initial legislation often serves as a foundational step that requires subsequent adjustments. For instance, if new AI techniques emerge that circumvent the current definitions of "digitization" or "photorealistic image," further legislative action may be required. The ongoing dialogue between technology experts, legal scholars, civil liberties advocates, and victim advocates will be essential in informing these evaluations. The goal of such monitoring would be to ensure that the law remains robust, adaptable, and effective in protecting individuals from the harms of non-consensual deepfakes, without inadvertently stifling legitimate technological innovation or artistic expression. Public feedback and case outcomes will serve as critical inputs for any future policy adjustments or enhancements to the regulatory framework.

Penalties, Liability, and Appeals

Under California SB 926, the creation and distribution of non-consensual AI-generated sexually explicit images is classified as a misdemeanor offense. This classification carries significant legal consequences for perpetrators, typically involving potential penalties such as imprisonment in a county jail for up to one year, substantial fines, or both. The specific sentencing will depend on various factors, including the severity of the offense, the emotional distress caused to the victim, and the offender's prior criminal history. The intent element is crucial for establishing liability: the prosecution must prove that the individual intentionally created and/or distributed the image, and that they knew or should have known that its distribution would cause serious emotional distress to the depicted person. This "knew or should have known" standard introduces an objective component, meaning that even if an offender claims ignorance, they can still be held liable if a reasonable person in their position would have understood the potential for harm.

The bill also ensures that victims have avenues for redress beyond criminal prosecution. While SB 926 primarily establishes criminal penalties, victims may also pursue civil remedies, such as filing civil lawsuits for damages, including emotional distress, reputational harm, and other losses incurred as a direct result of the deepfake's creation and distribution. This dual approach of criminal and civil recourse provides a more comprehensive framework for justice. In terms of appeals, individuals convicted under SB 926 will have the right to appeal their conviction and/or sentence through the standard appellate process within the California judicial system. This right to appeal ensures due process and allows for a review of legal errors or factual disputes that may have occurred during the trial. The availability of both criminal penalties and civil avenues for compensation aims to provide robust protection for victims and strong deterrence against the malicious use of AI to generate sexually explicit deepfakes.

Relationship to Other Instruments

California SB 926 does not operate in isolation but is intricately linked to and builds upon existing legal instruments, most notably California Penal Code Section 647(j)(4), which addresses the non-consensual distribution of intimate images, often referred to as "revenge porn" laws. Prior to SB 926, these laws primarily covered the distribution of actual photographs or videos. SB 926 specifically expands this existing framework by explicitly including images created or altered through "digitization," thereby encompassing AI-generated deepfakes. This amendment closes a critical loophole that emerged with the advancement of generative AI, ensuring that the legal protections against non-consensual intimate imagery extend to fabricated content that appears authentic. By amending an existing code section rather than creating an entirely new statute, SB 926 leverages established legal interpretations and enforcement mechanisms, facilitating its integration into the state's criminal justice system.

Furthermore, SB 926 is part of a broader package of AI-related legislation enacted in California during the 2024 legislative session, signaling a comprehensive effort by the state to regulate artificial intelligence. Other related instruments include SB 981, which requires social media platforms to establish reporting mechanisms for sexually explicit deepfakes and to promptly investigate and remove such content. Another relevant bill is AB 1831, which expands child pornography laws to include AI-generated child sexual abuse material. While SB 926 focuses on the criminalization of creating and distributing adult non-consensual deepfakes, SB 981 addresses the platform responsibility aspect, and AB 1831 tackles the particularly heinous crime involving minors. Together, these laws form a multi-faceted regulatory approach to combat the misuse of AI in generating harmful sexual content, demonstrating a coordinated legislative strategy to protect individuals from various forms of digital exploitation and harassment. The interplay between these instruments creates a more holistic legal environment for addressing the challenges posed by advanced AI technologies.

International Alignment

California's enactment of SB 926 places it among a growing number of jurisdictions globally that are grappling with the legal and ethical challenges posed by artificial intelligence, particularly in the context of deepfakes and non-consensual intimate imagery. While SB 926 is a state-level law, its underlying principles resonate with international discussions and legislative efforts aimed at protecting individuals from digital manipulation and exploitation. Many countries and international bodies, such as the European Union with its AI Act, are exploring or have implemented regulations that address the responsible development and deployment of AI, including provisions related to transparency, accountability, and the prevention of harm. The specific focus of SB 926 on criminalizing non-consensual deepfake pornography aligns with a broader global recognition of the severe psychological and reputational damage these fabricated images inflict, and the urgent need for legal recourse for victims.

Although there isn't a direct international treaty or convention that SB 926 explicitly aligns with, the bill reflects a common concern shared by legal systems worldwide regarding the misuse of advanced technology to violate fundamental rights, including privacy, dignity, and personal security. Jurisdictions like the United Kingdom, several Australian states, and other countries have also introduced or are considering legislation to criminalize deepfake pornography, often building upon existing laws against revenge porn or harassment. California's approach, which amends existing criminal statutes to incorporate AI-generated content, provides a model for other regions seeking to adapt their legal frameworks to new technological realities. The global nature of digital content distribution also implies that while SB 926 provides legal protection within California, cross-border cooperation and harmonization of laws will be increasingly important to effectively combat the international spread of non-consensual deepfakes. This bill contributes to the evolving international legal landscape that seeks to establish clear boundaries for the ethical use of AI and to hold perpetrators of AI-enabled harm accountable.

Implementation Timeline

Compliance Checklist

Sources and References