California Strengthens AI Regulation: Key Developments from the 2026 Legislative Session

This blog was originally posted on 6th November, 2026. Further regulatory developments may have occurred after publication. To keep up-to-date with the latest compliance news, sign up to our newsletter.

AUTHORED BY SAMANTHA ANGUIANO, Regulatory Compliance Specialist, ADHERENT

---

Key Insight:  California’s 2026 legislative session brought a new set of laws governing how businesses develop and use artificial intelligence (AI). The legislation addresses AI-generated content, advertising, customer service, children’s online safety, and the workplace. Rather than establishing a single AI regulatory framework, California adopted targeted requirements for different technologies and business activities. The new laws affect both AI providers and businesses that use AI, with several requirements taking effect in 2027.

Table of Contents

• What California Passed
• AI in Products and Digital Services
• AI in Advertising
• AI in Customer Service
• AI in the Workplace
• Key Takeaways
• Frequently Asked Questions

What California Passed

During its 2026 legislative session, California enacted nine laws addressing the use of AI and related technologies across multiple sectors. These measures introduce disclosure requirements, protections for children, restrictions on workplace surveillance, and limits on automated employment decisions.

It is important to note that these laws do not apply uniformly to every business using AI. Their scope depends on the technology involved, the activities conducted, and the entities covered. The following sections examine the legislation by the business activities it affects: products and digital services, advertising, customer service, and the workplace.

AI in Products and Digital Services

Three major enacted laws address AI systems and digital services, with particular attention to children’s safety, privacy, and transparency.

Adam’s Law Sets Child Safety Rules for Companion Chatbots

Senate Bill 1119, known as Adam’s Law, establishes additional safeguards for children interacting with AI companion chatbots. Operators that make companion chatbots available in California must determine users’ ages or apply the required child protections to all users.

For chatbots accessible to children, the law introduces several requirements:

• Risk assessments: Operators must assess and mitigate risks to children before making a new or substantially modified companion chatbot available and publish a child safety policy.
• Protective default settings: Operators must disable persistent conversational memory for child users, restrict push notifications, and set session limits of one hour and daily use limits of two hours, subject to parental controls.
• Crisis response: Operators must maintain protocols addressing suicidal ideation and self-harm, including referrals to crisis services and additional responses when an imminent risk is identified.
• Prohibited chatbot behavior: Operators must take reasonable measures to prevent chatbots from claiming to be human or sentient, expressing romantic interest in children, encouraging self-harm or emotional dependence, or soliciting purchases to maintain a relationship.
• Data protection and advertising: The law restricts targeted advertising to children and the sale of their personal information.
• Independent audits: Covered operators must undergo independent child safety audits and comply with related reporting requirements.

The legislation establishes phased implementation dates. Certain child safety obligations begin July 1, 2027, while additional requirements apply in subsequent years. Initial independent child safety audits are generally required by January 1, 2029, or before the operator first makes a companion chatbot publicly available, whichever is later. Operators with less than $500 million in gross revenue in the preceding calendar year are temporarily exempt from the audit requirements until January 1, 2032.

California Replaces Its Age-Appropriate Design Code

Assembly Bill 2246 repeals and replaces the California Age-Appropriate Design Code Act, establishing requirements for businesses offering online services, products, or features likely to be accessed by children under 18.

Whether a service is likely to be accessed by children depends on factors such as content directed toward minors, advertising aimed at children, features that appeal to younger audiences, and internal research indicating a significant child user base.

Covered businesses must estimate users’ ages or apply child protections to all users, establish high-privacy default settings, and provide children with a clear signal when someone is monitoring or tracking them.

The law also restricts profiling children, collecting their precise geolocation by default, and using dark patterns. In addition, businesses must take reasonable steps to prevent foreseeable physical, financial, and severe psychological harm associated with their services.

Although AB 2246 is not exclusively an AI law, its requirements are relevant to online services that incorporate AI-driven features, including profiling and personalized digital experiences.

California Expands Its AI Transparency Act

Senate Bill 100 amends the California AI Transparency Act (CATA), expanding its scope and updating requirements for identifying AI-generated content.

One of the principal changes is the removal of the previous threshold of one million monthly users or visitors. As a result, the definition of a covered provider now includes anyone who creates, codes, or otherwise produces a generative AI system that is publicly accessible in California, regardless of its number of users.

Covered providers are subject to requirements concerning:

• Latent disclosures: AI-generated or AI-altered image, video, and audio content must contain embedded information identifying the provider, the system and version, the creation date, and a unique identifier, as applicable.
• Disclosure verification tools: Providers must make a free tool available that allows users to assess whether digital content was created or altered by their generative AI systems.
• Licensee obligations: Providers must inform licensees of applicable requirements and take specified actions when modified versions of their systems no longer comply.
• Personal information: The law establishes restrictions on the collection, use, retention, and disclosure of personal information obtained through disclosure verification tools, subject to specified exceptions.

The legislation also establishes exemptions for certain technologies and temporarily excludes qualifying generative AI systems primarily designed as assistive technology.

AI in Advertising

California Requires Disclosure of AI-Generated Performers in Ads

Senate Bill 1050 establishes disclosure requirements for advertisements that prominently feature synthetic performers created using generative AI.

A synthetic performer is a digital figure, voice, or representation created wholly or partially through generative AI that resembles a human performance without depicting an identifiable individual.

A performer is considered prominent when it appears in the foreground demonstrating a product, delivers the narration or commercial message, or illustrates or reacts to that message.

Under the law, advertisements featuring these performers must include a clear and conspicuous disclosure indicating that the performance is synthetic. The disclosure must be substantially similar to “this performance features a synthetic performer” or “no human performer is depicted.”

The requirements apply to audio, video, and audiovisual advertisements, with responsibility placed on the person or entity that creates and publishes the advertisement rather than solely on the AI technology provider.

The legislation also establishes a process for removing advertisements found unlawful by a court. Once an advertising platform is served with the relevant court order, it must remove the advertisement and stop accepting payment for its distribution.

Certain advertisements promoting expressive works, including films, television series, and video games, are excluded when the synthetic performer is used as depicted in the underlying work. The law also provides exceptions for specified uses of AI involving translation or accessibility.

AI in Customer Service

Large Businesses Must Offer Access to a Human Agent

Assembly Bill 1609 establishes requirements for large private businesses that use AI chatbots to provide customer service or support to California customers.

The law applies to private businesses with more than $500 million in gross annual revenue nationally.

Covered businesses must comply with several requirements:

• AI disclosure: Customer service chatbots cannot be represented as human. When a reasonable person could be misled into believing they are interacting with a human, the business must clearly disclose that the chatbot uses AI through the same communication channel.
• Access to human representatives: During regular business hours, customer service channels must provide a straightforward way for customers to request assistance from a human representative through common commands, prompts, or menu options.
• Response times: Businesses must make a good-faith effort to connect customers with a human representative within 15 minutes or arrange a callback within one business day. Additional limits apply to subsequent hold times.
• Telephone support: Businesses that sell products or services online and offer telephone customer support must clearly display their customer service telephone number on their websites.

The law includes exceptions for certain communications, including internal and business-to-business support channels, specified healthcare communications, consumer reporting agencies, and circumstances beyond a business’s control.

Together, these requirements set clear standards for how covered businesses disclose AI use and provide access to human assistance in customer service interactions.

AI in the Workplace

Four major laws address the use of AI, automated decision systems, and workplace surveillance technologies in California. Each establishes requirements or restrictions for specific employment practices, ranging from employee monitoring and disciplinary decisions to notices of technology-related workforce displacement.

The No Robo Bosses Act Limits AI in Discipline and Termination

Senate Bill 947, known as the No Robo Bosses Act, regulates the use of automated decision systems (ADS) in employment decisions and takes effect on July 1, 2027.

The legislation defines an ADS as a computational process that uses technologies such as machine learning, statistical modeling, data analytics, or AI to produce outputs, including scores, classifications, or recommendations, that assist or replace human decision-making and materially affect individuals.

Under the law:

• Employers may not rely exclusively on an ADS to discipline or terminate an employee.
• When an employer primarily relies on an ADS for a disciplinary or termination decision, a human must corroborate the output using other relevant information, such as supervisor evaluations, personnel records, or work product.
• Employers must provide affected employees with written notice explaining the use of an ADS, the involvement of human review, and the appropriate contact for questions.
• Employees may request a description of their own data primarily used by the system in the decision.
• Employers may not use an ADS to infer an employee’s protected status or predict and penalize the exercise of legally protected rights.

The law also prohibits retaliation against employees who exercise their rights under these provisions.

California Restricts AI Emotion Recognition and Neural Data Collection at Work

Assembly Bill 1883 prohibits employers from using workplace surveillance tools that rely on AI to recognize or infer employees’ emotional states or to collect neural data.

The legislation addresses emerging monitoring technologies capable of analyzing sensitive information about employees beyond traditional workplace surveillance.

Exceptions apply to specified safety-related uses and circumstances in which monitoring is necessary to comply with federal requirements involving aerospace and defense activities.

California Restricts Workplace Surveillance in Bathrooms

Assembly Bill 1331 establishes restrictions on workplace surveillance by prohibiting employers from monitoring employees in workplace bathrooms using surveillance tools, whether or not those tools incorporate AI, unless authorized by a court order.

The legislation also allows employees to leave certain surveillance devices outside these areas.

Exceptions apply to specified identification and safety devices, including badges and personal alarms, provided they meet the law’s restrictions concerning audio and video recording and embedded AI capabilities.

Unlike AB 1883, which focuses on particular AI-enabled surveillance practices, AB 1331 addresses workplace monitoring more broadly.

California Requires Disclosure of Technology-Related Workforce Displacement

Senate Bill 951 amends the California Worker Adjustment and Retraining Notification (Cal/WARN) Act to address employment changes resulting from AI and other automated technologies.

When a covered mass layoff, relocation, or termination is caused wholly or substantially by technology replacing employee functions, the required notice must include additional information concerning the displacement.

This information includes:

• The number, job classifications, and locations of affected positions.
• The job functions that AI or other automated technologies will replace.
• The type of AI system or technology involved.
• A statement identifying the notice as one involving technological displacement.

The legislation also provides for the California Employment Development Department to publish summaries of these notices.

While SB 947 addresses the use of automated systems in individual disciplinary and termination decisions, SB 951 focuses on workforce changes associated with technological displacement.

Key Takeaways

California’s 2026 legislative session introduced new requirements governing the use of AI and related technologies across several business activities. The legislation addresses transparency in AI-generated content and customer service interactions, strengthens protections for children using digital services and companion chatbots, and establishes restrictions on automated employment decisions and workplace surveillance.

Rather than adopting a single comprehensive AI law, California has taken a targeted approach, regulating specific technologies and their use in different contexts. These developments are relevant to both AI providers and businesses that incorporate AI into their operations, although the scope and applicability of each law vary.

With some provisions already effective and others scheduled to take effect in 2027 or later, the legislation represents a significant development in California’s regulation of AI, automated technologies, and their impact on consumers and workers.

Frequently Asked Questions

Did California adopt a single comprehensive AI law in 2026?

No. Rather than adopting a single comprehensive AI law, California enacted nine laws addressing specific technologies and business activities, including AI-generated content, advertising, customer service, children’s online safety, and the workplace.

What protections did California introduce for children interacting with AI?

Senate Bill 1119 establishes safeguards for children using AI companion chatbots, including risk assessments, protective default settings, crisis response protocols, restrictions on certain chatbot behaviors, and independent child safety audits. AB 2246 establishes broader protections for online services, products, and features likely to be accessed by children under 18.

Are businesses required to disclose when customers interact with AI chatbots?

Under AB 1609, covered private businesses with more than $500 million in gross annual revenue nationally must clearly disclose AI use when a reasonable person could be misled into believing they are interacting with a human. The law also establishes requirements for access to human customer service representatives.

Can California employers rely exclusively on AI systems to discipline or terminate employees?

No. Under SB 947, employers may not rely exclusively on an automated decision system to discipline or terminate an employee. When an employer primarily relies on such a system, a human must corroborate its output using other relevant information. SB 947 takes effect on July 1, 2027.

See Adherent in Action

Discover how agentic AI is reshaping product compliance for global enterprises.

BOOK A DEMO