Skip to content
WaiterPilot™

Explainers

AI chatbot disclosure laws: do restaurants have to tell guests it's AI?

California, Utah, Maine and the EU have rules on telling people they are talking to AI. What each says, who it covers, and simple wording for your table card.

WaiterPilot Research
Updated , 10 min read

Abstract illustration of a glowing hologram ring

The short answer

In some places, yes. California bans bots that hide being a bot to push a sale, Maine requires notice when a chatbot could pass for a person, Utah requires an honest answer when asked, and the EU requires AI chat to be identified from 2 August 2026. The fix: call it an AI waiter, on the card and in chat.

Key points

  1. In the US the rules come state by state, and the EU AI Act's transparency rule for AI that talks to people applies from 2 August 2026.
  2. California's bot law, in force since 1 July 2019, protects a business that discloses its bot; Maine's 2025 law requires clear notice whenever a chatbot could make a reasonable consumer think they are talking to a person.
  3. Utah requires a business to say it is AI when a customer clearly asks, and gives a safe harbor to AI that says so at the start and throughout.
  4. A name for your AI waiter is fine, as long as the card and the first message say it is AI and a guest can always reach a person.
  5. This is a plain summary, not legal advice; laws change, so check the current text or ask a lawyer where you operate.
On this page

In some places, yes. California makes it unlawful to use a bot that hides being a bot to push a sale, Maine requires clear notice whenever a chatbot could pass for a person, Utah requires an honest answer when a customer asks, and the EU AI Act requires AI that talks to people to be identified from 2 August 2026. For a restaurant, the fix is the same everywhere and costs nothing: say it is an AI waiter on the table card and in the chat’s first line, and make a person easy to reach.

This is a plain summary, not legal advice. We quote each rule from the primary source we opened on 14 Nov 2026. Laws change and their reach depends on the facts, so check the current text or ask a lawyer where you operate.

Do I have to tell customers they are talking to AI?

Short answer: sometimes by law, and always by good sense.

We found no general US federal chatbot disclosure law. Instead, a few states have their own, written in different ways: one targets bots that mislead to make a sale, one requires notice whenever a guest could be fooled, and one requires an honest answer when asked. The EU has one rule for the whole bloc.

Guests care too. In a June 2026 survey of 1,000 US adults by PAR Technology, which sells restaurant software, 47% said AI goes too far if it replaces most human interaction, and 44% when they “cannot easily reach a person” (PAR, reported by QSR Web, 2026). A clear “I’m an AI waiter” with a waiter button next to it answers both worries.

Which chatbot disclosure laws could apply to a restaurant?

Short answer: these four, plus two that are often mentioned but don’t cover an ordering chat.

As read on each official source on 14 Nov 2026:

Place Law Who it covers What to disclose In force since
California SB 1001, Bus. & Prof. Code 17940 to 17943 Anyone using a bot to talk to people in California online, with intent to mislead about its artificial identity to push a sale or influence a vote That it is a bot; “clear, conspicuous” 1 Jul 2019
Utah SB 149 (2024), replaced by SB 226 (2025), Utah Code 13-75-103 Suppliers using generative AI with an individual in a consumer transaction; stricter for licensed professions That it is AI and not a human, if the customer clearly asks 1 May 2024; current rule 7 May 2025
Maine LD 1727, Public Law 2025, ch. 294, 10 MRSA §1500-DD Anyone using an AI chatbot in trade and commerce with a consumer Clear and conspicuous notice that the consumer is not dealing with a human, where a reasonable consumer could be misled 24 Sep 2025
European Union AI Act, Article 50(1) Providers of AI systems that interact directly with people That the person is interacting with AI, unless obvious; at the first interaction at the latest 2 Aug 2026
California SB 243, companion chatbots Companion chatbots; excludes bots used only for customer service or business operations Does not cover an ordering chat Chaptered 13 Oct 2025
Colorado SB 24-205, repealed and reenacted by SB 26-189 Automated decisions in areas such as jobs, housing, lending, insurance and health care Not aimed at restaurant ordering Signed 14 May 2026

We could not open New York’s official statute pages, so we leave New York out rather than guess.

What does California’s bot law require?

Short answer: don’t use a bot that hides being a bot to win a sale; disclosing it protects you.

California’s SB 1001 added sections 17940 to 17943 to the Business and Professions Code. It defines a bot as “an automated online account where all or substantially all of the actions or posts of that account are not the result of a person.” Section 17941 makes it unlawful to use a bot to talk to a person in California online “with the intent to mislead the other person about its artificial identity for the purpose of knowingly deceiving the person about the content of the communication in order to incentivize a purchase or sale of goods or services in a commercial transaction or to influence a vote in an election.”

The same section gives the way out: “A person using a bot shall not be liable under this section if the person discloses that it is a bot.” The disclosure must be “clear, conspicuous, and reasonably designed to inform persons with whom the bot communicates or interacts that it is a bot.” Section 17943 made it operative on 1 July 2019 (SB 1001 text).

Whether a restaurant’s ordering chat counts as an “online account” is a question for a lawyer. You don’t need the answer: a chat that takes orders is part of a sale, and a one-line disclosure puts you on the safe side.

What changed in Utah’s AI disclosure law in 2025?

Short answer: Utah narrowed the general rule to “tell the truth when asked” and added a safe harbor for AI that says so up front.

Utah’s SB 149, in effect from 1 May 2024, required anyone using generative AI in a matter its consumer protection division enforces to “clearly and conspicuously disclose” that it is AI and not a human, “if asked or prompted by the person” (SB 149).

SB 226 replaced that section from 7 May 2025 with Utah Code 13-75-103 (SB 226):

  • A supplier using generative AI with an individual in a consumer transaction must disclose that the individual is interacting with generative AI “and not a human” if the individual asks. The question must be “a clear and unambiguous request” to know.
  • Safe harbor: no enforcement action if the AI “clearly and conspicuously” discloses, “at the outset of any interaction” and “throughout the interaction”, that it is generative AI, is not human, or is “an artificial intelligence assistant.”
  • It is no defense that the AI “made the violative statement”. The business still answers for what its chat says.
  • The division can fine up to $2,500 per violation.

For a restaurant, the safe harbor is the practical target: say it at the start, and keep a label on screen.

What does Maine’s 2025 chatbot law say?

Short answer: if a reasonable guest could think the chat is a person, tell them clearly that it isn’t.

Maine’s law, now 10 MRSA §1500-DD, defines an AI chatbot as “a software application, web interface or computer program that simulates human conversation and interaction through textual or aural communications.” It says a person may not use one “to engage in trade and commerce with a consumer in a manner that may mislead or deceive a reasonable consumer into believing that the consumer is engaging with a human being unless the consumer is notified in a clear and conspicuous manner that the consumer is not engaging with a human being.” A violation is a violation of the Maine Unfair Trade Practices Act (10 MRSA §1500-DD).

The governor signed LD 1727 on 12 June 2025 as Public Law 2025, chapter 294 (LD 1727). The Legislature’s digest gives 24 September 2025 as the effective date for non-emergency laws from that session (Maine Legislature, 2025).

Does the EU AI Act apply to a restaurant chatbot?

Short answer: yes, from 2 August 2026, though the duty sits mainly with the company that provides the AI.

Article 50(1) says “Providers shall ensure that AI systems intended to interact directly with natural persons” are designed so “the natural persons concerned are informed that they are interacting with an AI system”, “unless this is obvious from the point of view of a natural person who is reasonably well-informed, observant and circumspect”. Article 50(5) says the information comes “at the latest at the time of the first interaction or exposure” (AI Act, Article 50).

On the date: the European Commission’s AI Act Service Desk, based on the consolidated text as at 27 July 2026, shows Article 113 saying the Act “shall apply from 2 August 2026”. The Digital Omnibus on AI moved several other dates, such as some high-risk rules, but Article 50 is not among the listed exceptions (AI Act, Article 113). A restaurant in the EU should ask its vendor how the chat meets Article 50, and still add its own line on the card.

What should the disclosure say? Wording you can copy

Short answer: what it is, what it answers from, and how to reach a person, in one or two lines.

Casa Olivo is a fictional restaurant we use for examples, not a customer. Its AI waiter is called Olivia.

Where Clear wording Wording to avoid
Table card “Scan to order with Olivia, our AI waiter. She answers from our menu. Want a person? Tap Call waiter or wave to us.” “Scan to chat with Olivia, your server tonight.”
Chat’s first line “Hi, I’m Olivia, Casa Olivo’s AI waiter. I answer from our menu. For allergies, please also tell our staff.” “Hi, I’m Olivia, I’ll be looking after you tonight!”
“Are you a real person?” “No, I’m an AI waiter. Tap Call waiter and one of our team will come over.” Any answer that dodges the question
Screen label “Olivia · AI waiter” kept in view A name alone, with no sign it is AI

The left column meets the spirit of all four rules: it says AI at the start, it stays visible, it answers honestly when asked, and it gives a path to a person.

Can you give your AI waiter a name?

Short answer: yes, as long as the name never stands in for a person.

A name gives the chat a friendly voice, and none of the laws above bans one. The risk is a name plus wording that suggests a human, such as “your server tonight”. Pair the name with “AI waiter” every time it appears.

With WaiterPilot you can give your AI waiter a name and a house style. Those settings shape tone only: they never change the menu, prices or what a guest can order. They also don’t make it a person, and your card and first line should say so.

How does this work with WaiterPilot? A 6-step checklist

Short answer: WaiterPilot is presented as an AI waiter, and you add the disclosure line yourself and check it.

We don’t claim WaiterPilot shows a legal notice on its own. Here is what to do, in order:

  1. Print the line on your QR cards. Each table has its own QR code with a Print QR codes button, and cards print in three sizes. Add the table-card wording above.
  2. Put the same line in the waiter’s welcome. Ask us during the free setup, then open your table’s chat on your own phone and read the first message.
  3. Ask it “are you a real person?” It should say it is an AI waiter. If not, tell us before you go live.
  4. Show the way to a person. Guests can call a waiter from their phone, and the call shows on the floor screen within seconds. Mention it on the card.
  5. Keep it on the menu. WaiterPilot answers from your menu and sends guests to staff when the menu doesn’t say. It steers medical or legal questions back to food, and guests cannot change its rules.
  6. Keep allergies with staff. It gives what the menu says and asks the guest to confirm with staff, who always have the final word.

For what happens when it misreads an order, see when the AI waiter gets it wrong. For how the three kinds of “AI waiter” differ, see what is an AI waiter?, and for what guests share, privacy and data.

When is WaiterPilot not the right pick?

Short answer: when you need a vendor that files compliance paperwork for you.

WaiterPilot has not published security certifications or a data processing agreement. If your legal team needs a vendor to certify its AI notices or sign compliance documents before launch, ask us in writing first, or choose a vendor that publishes them. And if your guests mostly want a person at the table, keep it that way; our look at whether AI will replace waiters explains why staff still make and serve the food.

What to do next

  1. Write your table-card line and the chat’s first line from the table above.
  2. Test the chat on your own phone: the first message, and “are you a real person?”
  3. If you operate in California, Utah, Maine or the EU, have a lawyer read your wording once.

Read more on the AI waiter page. WaiterPilot costs $200 per location per month after 30 days free, or $2,000 per location per year, with 0% commission, free setup and no contract; see pricing. Start 30 days free or book a demo.

Questions readers ask

Do I have to tell customers they are talking to AI?

In some places, yes. Maine requires clear notice when a chatbot could make a reasonable consumer think it is a person, Utah requires an honest answer when a customer clearly asks, and the EU requires AI that talks to people to be identified from 2 August 2026. Elsewhere it is still the safest practice.

What is the California chatbot disclosure law?

SB 1001, Business and Professions Code sections 17940 to 17943, in force since 1 July 2019. It makes it unlawful to use a bot to mislead a person in California about its artificial identity in order to push a sale or influence a vote, and a business that discloses its bot is not liable under it.

What are the AI chatbot disclosure requirements in Utah?

Since 7 May 2025, a supplier using generative AI with a customer must say it is AI and not a human if the customer clearly asks. AI that clearly says so at the start of the conversation and throughout gets a safe harbor. Licensed professions have stricter rules for high-risk conversations.

Does the EU AI Act require chatbot disclosure?

Yes. Article 50 requires providers to design AI systems that talk directly with people so those people are told they are dealing with AI, unless it is obvious, at the latest at the first interaction. The European Commission's official text gives 2 August 2026 as the date the Act applies.

Can I give my restaurant's AI chatbot a human name?

Yes, a name is fine, but don't let it pretend to be a person. Pair the name with the words AI waiter on the table card and in the first message, and make sure it answers honestly if a guest asks whether they are talking to a person.

What should a chatbot disclosure say?

Keep it short and plain: what it is, what it answers from, and how to reach a person. For example: Hi, I'm Olivia, Casa Olivo's AI waiter. I answer from our menu. Tap Call waiter any time to reach our staff.

Sources

  1. SB 1001: Bots: disclosure (Chapter 892, Statutes of 2018), CalMatters Digital Democracy (California Legislature bill text), 2018-09-28 (checked 2026-11-14)
  2. S.B. 149 Artificial Intelligence Amendments, enrolled copy, Utah State Legislature, 2024 (checked 2026-11-14)
  3. S.B. 226 Artificial Intelligence Consumer Protection Amendments, enrolled copy, Utah State Legislature, 2025 (checked 2026-11-14)
  4. 10 MRSA §1500-DD, Required disclosure of use of artificial intelligence chatbot to engage in trade and commerce, Maine Legislature, Office of the Revisor of Statutes, 2025 (checked 2026-11-14)
  5. Summary of LD 1727 (HP 1154), Maine Legislature, 2025-06-12 (checked 2026-11-14)
  6. Legislative digest of bills and summaries of enacted laws, 132nd Legislature, first regular and first special sessions, Maine Legislature, 2025 (checked 2026-11-14)
  7. AI Act, Article 50: Transparency obligations for providers and deployers of certain AI systems, European Commission, AI Act Service Desk, 2026-07-27 (checked 2026-11-14)
  8. AI Act, Article 113: Entry into force and application, European Commission, AI Act Service Desk, 2026-07-27 (checked 2026-11-14)
  9. SB 243: Companion chatbots (Chapter 677, Statutes of 2025), CalMatters Digital Democracy (California Legislature bill text), 2025-10-13 (checked 2026-11-14)
  10. SB26-189 Automated Decision-Making Technology, Colorado General Assembly, 2026-05-14 (checked 2026-11-14)
  11. 3 in 4 diners open to restaurant AI, PAR survey finds, QSR Web, 2026-08-24 (checked 2026-11-14)

Watch an order go from table to kitchen

Try a sample order on your phone, or book a 20-minute demo. 30 days free, no setup fee, no contract.