California's Bot Disclosure Law: When a Business Must Say a Chatbot Is Not Human

Last updated: November 2026. Reviewed by Pavel Kolmogorov, California business litigation attorney (State Bar No. 321018).

California's bot disclosure law makes it unlawful to use a bot online to mislead a person in California about its artificial identity, in order to knowingly deceive that person and incentivize a purchase. (Bus. & Prof. Code, § 17941, subd. (a).) A business that discloses the bot is not liable under that section. A sales chat agent should therefore say, clearly and at the start, that it is automated. Kolmogorov Law, P.C. advises businesses on AI compliance and litigates the disputes that follow.

Call (909) 235-6116 or contact us online to discuss a chatbot, an AI sales agent, or a demand letter about one.

What section 17941 prohibits, element by element

The prohibition has five parts.

A person uses a bot. "Person" includes corporations, limited liability companies, and partnerships.
The bot communicates or interacts with another person in California online.
The user intends to mislead that person about the bot's artificial identity.
The purpose is to knowingly deceive the person about the content of the communication.
The aim is to incentivize a purchase or sale of goods or services in a commercial transaction, or to influence a vote in an election.

(Bus. & Prof. Code, §§ 17940, subd. (d), 17941, subd. (a).)

A bot is an automated online account where all or substantially all of the actions or posts are not the result of a person. (Bus. & Prof. Code, § 17940, subd. (a).) The definition does not mention artificial intelligence, and it turns on the word "account." The statute does not say whether a chat tool on a company's own website is such an account. The cautious course is to treat it as one.

The chapter became operative on July 1, 2019. (Bus. & Prof. Code, § 17943.) As of November 2026 the text has not been amended.

This is not a general duty to label every chatbot. Liability requires intent to mislead and a purpose of knowingly deceiving, and a person who discloses the bot is not liable under the section.

The 10,000,000 visitor or user figure does not limit who is covered

An "online platform" is a public-facing site or application with 10,000,000 or more unique monthly United States visitors or users for a majority of months during the preceding 12 months. (Bus. & Prof. Code, § 17940, subd. (c).)

That term appears in one place. The chapter imposes no duty on service providers of online platforms, such as web hosts and Internet service providers. (Bus. & Prof. Code, § 17942, subd. (c).) The prohibition itself uses the word "online," which means appearing on any public-facing website, web application, or digital application. (Bus. & Prof. Code, § 17940, subd. (b).) The figure does not take a small company's website out of the statute.

What a compliant disclosure looks like

The disclosure must be clear, conspicuous, and reasonably designed to inform the people the bot communicates with that it is a bot. (Bus. & Prof. Code, § 17941, subd. (b).) The statute prescribes no wording. In practice:

Say it in the first message, before the customer relies on anything.
Keep a visible label in the chat window.
Do not rely on a line in the terms of use.

Other California rules that can reach an AI agent

Companion chatbots. A chapter operative January 1, 2026 regulates operators of companion chatbot platforms. If a reasonable person would be misled into believing the chatbot is human, the operator must give a clear and conspicuous notice that it is artificially generated. (Bus. & Prof. Code, § 22602, subd. (a).) A person injured by a violation may sue for an injunction, the greater of actual damages or $1,000 per violation, and attorney's fees. (Bus. & Prof. Code, § 22605.) The definition excludes a bot used only for customer service, a business's operational purposes, productivity and analysis related to source information, internal research, or technical assistance. (Bus. & Prof. Code, § 22601, subd. (b)(2)(A).) A bot designed to sustain a personal relationship with users may fall inside it. Two laws signed on September 10, 2026 change these rules on January 1, 2027. One prohibits making or selling a toy that includes a companion chatbot, or offering one to a retailer, until 2031. (Bus. & Prof. Code, § 22604.5.) The other adds child-safety duties for operators, most of them operative July 1, 2027. (Bus. & Prof. Code, §§ 21811, 21812.)

Telephone calls. Before an automatic dialing-announcing device plays a prerecorded message, a live announcement must tell the person called if the message uses an artificial voice. (Pub. Util. Code, § 2874, subd. (a)(3).)

Health care. A clinic or physician's office that uses generative AI for patient communications about clinical information must include a disclaimer and instructions for reaching a person, unless a licensed provider read and reviewed the message. (Health & Saf. Code, § 1339.75.) Since January 1, 2026, the advertising or functionality of an AI system may not use terms implying that its care or advice comes from a natural person licensed as a health care professional. (Bus. & Prof. Code, § 4999.9, subd. (c).)

Generative AI providers and automated decisions have their own rules, covered on our AI Transparency Act page and our page on the automated decisionmaking regulations.

Who enforces the California bot disclosure law

The bot chapter states no penalty or remedy of its own. Its duties are cumulative with other law. (Bus. & Prof. Code, § 17942, subd. (a).)

The Unfair Competition Law reaches any unlawful business act or practice. (Bus. & Prof. Code, § 17200.) A claimant can argue that a bot chapter violation qualifies, but the research for this page found no published decision deciding the point. Under that law, public prosecutors may seek civil penalties of up to $2,500 for each violation. A private plaintiff must have suffered injury in fact and lost money or property. (Bus. & Prof. Code, §§ 17204, 17206, subd. (a).)

What the bot says is a separate exposure. A defendant that used AI may not assert as a defense that the AI autonomously caused the harm, as our page on Civil Code section 1714.46 explains.

How it comes up for a business

A retailer gives its chat agent a first name and a staff photo. The agent answers product questions and offers a discount to close. A customer later says she believed she was dealing with a salesperson.

A lead-generation vendor runs automated social media accounts that message prospects as "advisors" for a client's service. Both companies may be said to use the bot, so the vendor contract should say who is responsible for disclosure.

Chat transcripts raise privacy claims of their own, discussed on our California Invasion of Privacy Act page.

What to bring to the first meeting

A list of every chat agent and automated account, including vendors' tools.
Screenshots of what a customer sees in the first message.
The vendor agreements and configuration settings.
Any complaint or demand letter received.

Frequently asked questions

Our website uses an AI chat agent for sales. Does California require us to tell customers they are talking to a bot?

Not in every case, but disclosure is the safe course. California makes it unlawful to use a bot online to mislead a person about its artificial identity in order to knowingly deceive that person and incentivize a purchase. A business that clearly and conspicuously discloses the bot is not liable under that section. The statute sets no timing, so the first message is the prudent place.

We are a small company with a few thousand site visitors a month. Does the California bot law still apply to us?

Your size does not exempt you. The 10,000,000 monthly visitor or user figure in the California statute defines an online platform, and that term is used only to exempt platform service providers such as web hosts. The prohibition applies to any person who uses a bot online, which includes any public-facing website or application.

What is a bot under California law?

California Business and Professions Code section 17940 defines a bot as an automated online account where all or substantially all of the actions or posts are not the result of a person. The definition does not mention artificial intelligence, so it is not limited to AI tools. The statute does not say whether a chat tool on a business's own website counts as an account.

What is the penalty for violating the California bot disclosure law?

The bot chapter states no penalty or remedy of its own. A claimant may argue that a violation is an unlawful business practice under California's Unfair Competition Law, which allows injunctions, restitution, and, in public actions, civil penalties of up to $2,500 per violation. The research for this page found no published decision deciding whether that law applies to the bot chapter.

Does the companion chatbot law apply to a customer service bot?

Generally not. California's companion chatbot chapter, operative January 1, 2026, excludes a bot used only for customer service, a business's operational purposes, productivity and analysis related to source information, internal research, or technical assistance. A bot built to sustain a personal relationship with users may be covered, and a covered operator faces notice duties and private suits.

Talk to a California AI compliance and business litigation attorney

If your business uses a chatbot or AI agent with customers, or has received a complaint about one, call (909) 235-6116 or contact us online. Kolmogorov Law, P.C. is in Irvine, California.

Related reading: our California AI compliance guide and business litigation practice.

This page provides general information about California law and is not legal advice. AI law is changing quickly, and outcomes depend on the facts, the current text of each statute, and the court.