United States - California - Chatbot Disclosure Requirements (SB 243)

Overview

California Senate Bill 243 (SB 243), known as the Companion Chatbot Disclosure Requirements, represents a landmark legislative effort in the United States, establishing the nation's first comprehensive state law specifically regulating the operation of AI-powered companion chatbots. Signed into law by Governor Gavin Newsom on October 13, 2025, and becoming effective on January 1, 2026, this legislation was enacted in direct response to growing public health concerns and several high-profile incidents involving minors who allegedly experienced self-harm or suicidal ideation linked to interactions with conversational AI systems. The primary objective of SB 243 is to safeguard users, particularly minors and vulnerable individuals, from potential psychological and emotional harms that can arise from engaging with AI chatbots designed to simulate human-like social and emotional interactions.

The law introduces a new regulatory baseline for the companion AI industry by imposing affirmative duties across three critical areas: mandatory disclosures, robust safety protocols, and enhanced accountability mechanisms. It mandates clear and conspicuous notifications to users about the artificial nature of their interactions, establishes requirements for operators to implement crisis prevention protocols for suicidal ideation and self-harm, and includes specific content guardrails to protect minors from exposure to inappropriate material. Furthermore, SB 243 empowers individuals with a private right of action, allowing them to seek legal recourse for injuries sustained due to non-compliance. This pioneering legislation positions California as a significant trendsetter in AI governance, with its provisions potentially influencing future national and international regulatory frameworks for artificial intelligence.

Definitions

California SB 243 introduces specific definitions to delineate the scope of its application, primarily focusing on the concept of a "companion chatbot" and the responsibilities of its "operator." A "companion chatbot" is defined as an artificial intelligence program or system designed to simulate conversation and companionship, including by expressing or imitating emotions or affection, thereby creating a human-like social experience for users. This definition is crucial as it distinguishes these systems from other AI applications, such as customer service bots, video game characters with limited dialogue, or voice-activated virtual assistants that do not aim to foster sustained emotional relationships or personalized interactions. The law targets AI systems that build ongoing emotional relationships with users, recognizing the unique ethical and psychological challenges they present.

An "operator" refers to any entity that makes a companion chatbot platform available to users in California. This broad definition ensures that both developers and providers of companion chatbot services are subject to the law's requirements, regardless of where the AI system was developed, as long as it is accessible to California residents. The legislation also places particular emphasis on "minors," defined as users whom the operator knows to be under a certain age (typically 18 in California law, though specific age may be clarified in implementing regulations if not explicitly stated in the bill text). The distinction between general users and minors is central to many of the bill's protective measures, mandating heightened disclosure and safety protocols when interactions involve underage individuals.

Governance and Institutional Framework

The governance and institutional framework for California SB 243 primarily designates the Office of Suicide Prevention within the State Department of Health Care Services (formerly State Department of Public Health for the Office of Suicide Prevention) as a key oversight body for specific aspects of the law. Beginning July 1, 2027, operators of companion chatbot platforms are required to submit annual reports to this office. These reports must detail the number of times crisis service provider referral notifications were issued in the preceding calendar year, outline the protocols established to detect, remove, and respond to instances of suicidal ideation by users, and describe protocols implemented to prohibit companion chatbot responses that encourage or facilitate suicidal ideation or actions. The Office of Suicide Prevention is further mandated to post data derived from these annual reports on its internet website, promoting transparency and public accountability regarding the mental health impacts of companion chatbots.

While the Office of Suicide Prevention plays a crucial role in data collection and public reporting related to mental health safeguards, the broader enforcement and compliance mechanisms for SB 243 are also supported by the provision of a private right of action. This allows individuals who suffer injury in fact due to an operator's non-compliance to bring civil actions, effectively empowering citizens to contribute to the enforcement of the law. Although some early discussions or proposals might have suggested oversight by other entities like the California Department of Technology, the enacted legislation clearly delineates specific reporting responsibilities to the Office of Suicide Prevention, underscoring the state's focus on addressing the psychological safety and well-being of users.

Key Focus Areas

California SB 243 establishes several key focus areas designed to enhance transparency, ensure user safety, and promote accountability within the companion chatbot industry. A central pillar of the legislation is its disclosure requirements. For general users, if a reasonable person could be misled into believing they are interacting with a human, operators must issue a clear and conspicuous notification stating that the companion chatbot is artificially generated and not human. For users known to be minors, operators must explicitly disclose that the user is interacting with artificial intelligence and provide clear and conspicuous notifications at least every three hours during ongoing interactions, reminding them to take a break and reiterating the chatbot's AI nature. Additionally, all platforms must include a suitability warning, disclosing that companion chatbots may not be appropriate for some minors.

Another critical focus is on safety protocols and content guardrails. Operators are mandated to maintain a robust protocol for preventing the chatbot from producing suicidal ideation, suicide, or self-harm content to the user. This protocol must include providing notifications that refer at-risk users to crisis service providers, such as suicide hotlines or crisis text lines, when expressions of suicidal ideation or self-harm are detected. Details of this crisis prevention protocol must be published on the operator's website. Furthermore, operators must implement reasonable measures to prevent chatbots from generating visual material of sexually explicit conduct or directly encouraging minors to engage in such conduct. The law also requires operators to use evidence-based methods for measuring suicidal ideation and to take reasonable steps to prevent chatbots from providing unpredictable rewards or encouraging increased engagement, usage, or response rates, especially for minors.

Implementation Framework

The implementation framework for California SB 243 outlines the steps and obligations for operators of companion chatbot platforms to achieve compliance with the new regulations. Upon the law's effective date of January 1, 2026, companies operating companion chatbot platforms are required to undertake an initial compliance assessment. This assessment involves a careful review to determine whether their AI systems fall within the statutory definition of a "companion chatbot" or qualify for any statutory exclusions, such as customer service bots or limited-dialogue video game characters. Understanding the scope of application is fundamental to initiating the compliance process.

A significant aspect of the implementation involves establishing and maintaining robust internal systems and processes. Operators must revise and update their disclosure mechanisms to ensure clear and conspicuous notifications are provided to both general users and minors, as mandated by the law. This includes developing systems to track interaction durations for minors to trigger periodic break reminders. Furthermore, companies must implement and publish comprehensive crisis prevention protocols on their websites, detailing how they detect, prevent, and respond to suicidal ideation or self-harm content, including the provision of crisis service referrals. Data collection and tracking systems must also be established to accurately capture and report the required metrics to the Office of Suicide Prevention, beginning July 1, 2027, while ensuring that no personal identifiers are included in these reports to maintain user privacy. Some legislative drafts also mentioned requirements for regular third-party audits to ensure compliance, emphasizing a proactive approach to regulatory adherence.

Monitoring and Evaluation

Monitoring and evaluation under California SB 243 are primarily centered around mandatory annual reporting requirements designed to assess the effectiveness of the implemented safety protocols, particularly concerning user mental health. Beginning July 1, 2027, operators of companion chatbot platforms are obligated to submit annual reports to the California Office of Suicide Prevention. These reports serve as a critical mechanism for the state to monitor the real-world impact of companion chatbots and evaluate the industry's adherence to the law's protective measures.

The annual reports must include specific data points and information. Operators are required to report the total number of times they have issued crisis service provider referral notifications in the preceding calendar year, indicating the frequency with which at-risk users were directed to mental health support resources. Additionally, the reports must detail the specific protocols that have been put in place by the operator to detect, remove, and respond to instances of suicidal ideation by users. This includes outlining the methodologies used to identify such expressions and the actions taken to mitigate harm. Furthermore, operators must describe the protocols established to prohibit a companion chatbot from generating responses that encourage or facilitate suicidal ideation or actions with the user. The Office of Suicide Prevention is then responsible for posting aggregated and anonymized data from these reports on its internet website, fostering public transparency and allowing for broader evaluation of the law's impact on public safety and mental well-being.

Penalties, Liability, and Appeals

California SB 243 introduces significant provisions regarding penalties, liability, and appeals, primarily through the establishment of a private right of action for individuals harmed by non-compliant companion chatbot operators. This mechanism empowers any person who suffers an "injury in fact" as a direct result of an operator's failure to comply with the requirements of the bill to bring a civil action against the offending party. This private right of action is a crucial enforcement tool, allowing individuals to seek redress for damages and compelling operators to adhere strictly to the law's mandates.

Under this provision, injured individuals may seek various forms of relief, including damages of at least $1,000 per violation, injunctive relief to prevent further non-compliance, and the recovery of attorney's fees. The inclusion of a minimum damage amount and the ability to recover legal costs are intended to make it more feasible for individuals to pursue claims, even for harms that might be difficult to quantify monetarily but nonetheless represent a significant violation of their rights under the law. This framework places a substantial burden of accountability on chatbot developers and operators, encouraging proactive compliance to avoid potential litigation and financial penalties. The law's strong liability provisions underscore California's commitment to protecting users from the potential harms associated with unregulated AI companion chatbots, particularly those affecting minors and vulnerable populations.

Relationship to Other Instruments

California SB 243 operates within a rapidly evolving landscape of AI regulation, both domestically and internationally, and its relationship to other legal instruments is noteworthy. While it stands as the first state law in the U.S. specifically targeting companion chatbots and their interactions with minors, it complements a broader wave of state-level legislative activity focused on AI governance. For instance, other states like New York, Utah, and Texas have also considered or enacted bills addressing various aspects of chatbot regulation, though SB 243 is distinct in its tailored protections for minors and its focus on emotional and psychological safety.

Notably, SB 243 emerged as the preferred legislative vehicle for addressing chatbot harms after Governor Newsom vetoed another related bill, Assembly Bill 1064 (AB 1064), which proposed more substantial limits on chatbots. This indicates a deliberate legislative choice to advance the specific disclosure and safety-focused framework of SB 243. While the bill does not directly amend or reference existing federal laws like Section 230 of the Communications Decency Act or broader state privacy laws like the California Consumer Privacy Act (CCPA), its requirements regarding data handling for reporting purposes (e.g., ensuring no personal identifiers are included) align with general data protection principles. The legislation also sets a precedent that could influence the development of future federal AI policies, potentially serving as a blueprint for national standards as AI technology continues to advance and its societal impacts become clearer.

International Alignment

California SB 243, while a state-level initiative, holds significant implications for international alignment in AI governance, particularly given California's status as a global hub for technology and its history of setting de facto national and even international standards for technology regulation. The bill's focus on individual well-being, emotional integrity, and user autonomy, especially concerning minors, distinguishes it from other major international AI regulatory frameworks. For example, the European Union's AI Act primarily focuses on systemic risk classification and market conformity, while China's Generative AI Measures emphasize content control and societal stability. SB 243, by legally recognizing and regulating systems that simulate affection, intimacy, or empathy, pioneers the regulation of emotional intelligence in AI, an emerging field that is likely to influence global standards.

The legislation's emphasis on transparency, safety protocols, and accountability for companion chatbots means that international providers offering services in California will need to adapt their transparency, moderation, and oversight mechanisms to comply with these new requirements. This could lead to a harmonization of certain practices across different jurisdictions as companies seek to maintain a consistent approach to compliance. The bill's "psychological safety by design" principle, which mandates the design of systems to prevent manipulation, addiction, or emotional distress, complements broader ethical AI principles being discussed globally, such as those found in the EU AI Act and the GDPR's ethics-by-design notions. Consequently, SB 243 could serve as a model for other countries or international bodies considering how to address the unique challenges posed by emotionally engaging AI systems, thereby fostering a degree of international alignment in the regulation of this specific AI application.

Implementation Timeline

Sources and References