Which states have chatbot AI disclosure laws?

As of August 2026, the states with enacted laws that most directly affect an ordinary business chat widget are Maine (10 M.R.S. Section 1500-DD), Utah (AI Policy Act, SB 149 as amended by SB 226), California (Business and Professions Code Section 17941, the B.O.T. Act), and Colorado (HB 26-1263, the Chatbot Safety Act, effective January 1, 2027). Separately, New York, California, Oregon and Washington have enacted companion-chatbot laws that regulate AI companion and relationship apps rather than customer-service bots. More than a dozen additional states passed chatbot legislation during 2026, so the list is still moving.

Do companion chatbot laws apply to my customer service bot?

Usually not. Companion-chatbot laws such as California SB 243, New York General Business Law Article 47, Oregon SB 1546 and Washington HB 2225 target AI systems designed to sustain ongoing personal or emotional relationships with users, and they carry extra duties around minors, self-harm response protocols, and recurring reminders that the user is talking to AI. A support widget that answers questions about orders and shipping is generally outside that definition. It can still be covered by the general commercial disclosure laws in Maine, Utah, California and Colorado.

What is the penalty for not disclosing an AI chatbot?

Penalties are set per state and are typically enforced by the state attorney general under that state's unfair or deceptive trade practices act. Maine allows civil penalties of up to $1,000 per violation under the Maine Unfair Trade Practices Act. Colorado treats a violation of HB 26-1263 as a deceptive trade practice under the Colorado Consumer Protection Act, carrying civil penalties of up to $20,000 per violation. Some newer state laws, including Oregon SB 1546, also create a private right of action.

What should an AI chatbot disclosure actually say?

Keep it short, plain, and visible before the conversation starts rather than buried in a privacy policy or terms page. A line such as "You're chatting with an AI assistant. Ask any time to be connected to a person." satisfies the common core of the state laws: it identifies the system as AI up front, it is clear and conspicuous, and it gives the user a path to a human. Placing it inside the chat window as the first message, and in visible text near the launcher, covers both the always-on and on-request triggers.

Do I Have to Tell Customers My Chatbot Is AI? State-by-State Laws (2026)

Q: Do I have to tell customers my chatbot is AI?

It depends on your state and on what your chatbot does. There is no single federal rule. Maine requires a clear and conspicuous notice whenever an AI chatbot used in trade or commerce could mislead a reasonable consumer into thinking they are talking to a human. Utah requires disclosure when a consumer clearly and unambiguously asks whether they are talking to AI, and requires it up front in regulated professions and high-risk interactions. California prohibits using a bot to deceive someone in order to incentivize a sale or influence a vote unless the bot is disclosed. Colorado will require operators of public conversational AI services to disclose that the service is AI starting January 1, 2027. Because the triggers overlap and are expanding, most businesses find that a single always-on notice near the chat widget satisfies all of them.

There is no federal rule. Whether you must disclose depends on your state and on what your bot actually does. Here is every enacted U.S. chatbot disclosure law, what triggers it, and the one notice that satisfies almost all of them.

Two different kinds of chatbot law

Most articles on this topic produce a scary list of fifteen states. That list is misleading, because it mixes together two categories that have very different scopes:

General commercial disclosure laws — these can reach an ordinary business support widget. Maine, Utah, California and Colorado are the ones that matter here.
Companion-chatbot laws — these target AI built to sustain an ongoing personal or emotional relationship with a user, and they add duties around minors, self-harm response protocols, and repeated "you're talking to AI" reminders. If your bot answers questions about shipping, you are almost certainly not an AI companion.

Read the first group closely. Skim the second unless you build consumer AI companions.

General commercial chatbot disclosure laws

Maine — the broadest rule in the country

Maine's transparency act (LD 1727, codified at 10 M.R.S. §1500-DD) was signed June 12, 2025 and took effect September 16, 2025. It bars using an AI chatbot in trade or commerce in a way that may mislead or deceive a reasonable consumer into believing they are dealing with a human, unless the consumer is notified clearly and conspicuously that they are not. The definition of "chatbot" is deliberately broad — text interfaces and voice assistants alike. Violations are Maine Unfair Trade Practices Act violations, enforceable by the Attorney General, with civil penalties up to $1,000 per violation.

Practical read: this is the law most likely to catch a normal business. It has no "consequential decision" gate and no requirement that you intended to deceive — a convincingly human-sounding bot with no notice is the whole violation.

Utah — disclose when asked, and up front in regulated fields

Utah's AI Policy Act (SB 149) took effect May 1, 2024 and was narrowed by SB 226 in 2025. For ordinary consumer interactions, disclosure is now required only when a consumer makes a clear and unambiguous request to know whether they are talking to a human or AI. Proactive, up-front disclosure is still required in regulated occupations — the state-licensed professions — and in high-risk interactions involving health, financial or biometric data, or where the AI gives financial, legal or health care advice.

Utah also offers a safe harbor: if the AI itself clearly and conspicuously identifies as AI at the outset of and throughout the interaction, the business is not exposed on the disclosure requirement. That safe harbor is the single strongest argument for making your notice always-on rather than answer-on-request.

California — deception-triggered, plus a companion law

California's B.O.T. Act (Bus. & Prof. Code §17941, from SB 1001) is narrower than people assume. It makes it unlawful to use a bot to communicate with a person in California with intent to mislead about the bot's artificial identity in order to knowingly deceive them into a purchase or sale, or to influence a vote. Ordinary, non-deceptive automation is not covered. The statute has an express safe harbor: disclose clearly and conspicuously that it is a bot and there is no liability. California also enacted SB 243, a companion-chatbot law effective January 1, 2026.

Colorado — the strictest rule, arriving January 1, 2027

Colorado's HB 26-1263, the first-in-nation Chatbot Safety Act, was signed May 29, 2026 with duties applying January 1, 2027. Operators of publicly available conversational AI services must disclose that the service is AI and not a human, run a self-harm response protocol, take steps around minors, and report annually to the Attorney General. Violations are deceptive trade practices under the Colorado Consumer Protection Act — up to $20,000 per violation. Colorado's separate ADMT law, SB 26-189, takes effect the same day and adds notice duties when automation materially influences a consequential decision.

Full detail: the Colorado chatbot law guide and the Colorado SB 26-189 guide.

Companion-chatbot laws (probably not you)

These regulate AI companions rather than customer service. All of them require telling users, repeatedly, that they are interacting with AI:

New York — General Business Law Article 47, effective November 5, 2025.
California — SB 243, effective January 1, 2026.
Oregon — SB 1546, effective January 1, 2027, and notable for including a private right of action.
Washington — HB 2225, effective January 1, 2027.

A further group of states — including Connecticut, Georgia, Hawaii, Idaho, Iowa, Nebraska, Rhode Island, South Carolina and Wyoming — enacted chatbot legislation during 2026 with effective dates spread across 2026 and 2027. Scope and dates in that group are still being reported inconsistently, so confirm the current statutory text before relying on any summary, including this one.

The notice that satisfies all of them

The enacted laws differ on when you must disclose, but they agree on what an adequate disclosure looks like: clear, conspicuous, and reachable before the user has meaningfully engaged. One always-on notice clears every trigger above:

You're chatting with an AI assistant. Ask any time to be connected to a person.

Put it in visible text near the chat launcher, not only inside the window.
Make it the first message in the conversation as well — that is what Utah's safe harbor rewards.
Do not rely on a line in your privacy policy or terms. "Clear and conspicuous" means at the point of interaction.
Give a real path to a human. Several statutes assume one exists.

How to find out where you stand

Most businesses do not know which AI tools their own site loads — chat widgets get installed by a marketing agency or bundled into a theme and then forgotten. ComplianceBeacon scans your site, identifies the AI and chatbot scripts actually running on it, and tells you which state disclosure rules that combination triggers. Run a free scan.

This article provides educational information only, not legal advice. Statutory scope, effective dates and penalties change frequently and vary by fact pattern; consult a qualified attorney for compliance decisions specific to your business. Facts above were verified against state statutory sources and legal analyses as of August 3, 2026.

Do I Have to Tell Customers My Chatbot Is AI?