It depends on who you are, what you publish and where it lands. In the US there's no general rule that requires labeling AI-generated marketing, but there are specific ones: Texas health care providers must tell patients, and the FTC bans AI-generated fake reviews. In the EU, since August 2026, people must know when they're talking to an AI.
And there's a layer most teams miss: your AI vendor's usage policy, which can require disclosure even where the law doesn't.
What does each rule require?
As of October 2026:
| Where | What it requires | Who it applies to |
|---|---|---|
| US, federal | No general AI labeling rule. Advertising can't be deceptive, whoever or whatever wrote it, and banks and credit unions still follow Regulation DD's advertising rules | Everyone |
| FTC fake reviews rule | Bans fake reviews and testimonials, including AI-generated ones that pose as people who don't exist | Every business |
| Texas (TRAIGA) | Government agencies must tell consumers they're interacting with AI, and health care providers must tell patients when AI is used in their care | Government and health care providers |
| EU (AI Act, Art. 50) | Tell people they're interacting with AI; mark generated content in a machine-readable way; disclose deepfakes and AI-generated text on matters of public interest | Providers and professional users of AI whose systems or content reach people in the EU |
| Anthropic's Usage Policy (from November 12, 2026) | Consumer-facing chatbots must disclose they're AI, and people who receive high-risk recommendations made with AI (credit, insurance, health) must be told AI was used | Anyone using Claude for those cases |
What does TRAIGA require in Texas?
The Texas Responsible Artificial Intelligence Governance Act took effect January 1, 2026. Section 552.051 has two disclosure duties:
- Government agencies must tell consumers, before or at the time of the interaction, that they're interacting with an AI system, even if it would be obvious to a reasonable person.
- Health care providers using AI in a patient's care or treatment must tell the patient or their representative no later than the date the service is first provided, or as soon as reasonably possible in an emergency.
The disclosure must be clear and conspicuous, in plain language, and can't use a dark pattern. A hyperlink to a separate page is allowed. If you're a Texas hospital, clinic or practice, this one applies to you directly.
What does the EU require?
Article 50 of the AI Act has applied since August 2, 2026, and has four parts:
- Chatbots. People must know they're interacting with an AI system, unless it's obvious.
- Content marking. Providers of generative AI must mark generated text, images, audio and video in a machine-readable format. Systems already on the market before August 2, 2026 have until December 2, 2026.
- Deepfakes. Whoever publishes realistic generated or manipulated images, audio or video must disclose it.
- Public-interest text. Whoever publishes AI-generated text to inform the public on matters of public interest must disclose it, unless it went through human review and a person or company holds editorial responsibility.
If your organization serves customers or publishes content in the EU, check with counsel how it applies to you.
What does a practical rule look like?
While the laws catch up, this rule works anywhere:
- Always disclose when a customer is talking to AI. A chat assistant on your website or app introduces itself as a virtual assistant in its first message.
- Label realistic AI-generated images and video, especially if they show people.
- Never invent testimonials, reviews or customers.
- Disclose when AI played a part in a decision that affects a person, such as a loan, a claim or a health recommendation, and keep a person who reviews it in charge of the decision.
- For marketing copy reviewed and approved by a person, pick a policy and apply it the same way across channels. What matters is that the piece is accurate and someone is accountable for it.
An honest caveat: this area moves fast. Review the policy every quarter, because the EU published guidelines and a code of practice in 2026, and more states may follow Texas.
How do we handle it at Andes?
The AI policy in our governance package can cover when and how to disclose AI use. For the rest of the workflow, read who approves AI-generated marketing, the FDIC digital sign rule and NCUA ad statement and how to write a brand voice guide AI can follow.
Has your organization decided when it discloses AI use?
If every department decides on its own, you need one rule. See our AI for regulated institutions service and let's define it with your team.
Sources
- Texas Legislature, HB 149 (TRAIGA), section 552.051: https://capitol.texas.gov/tlodocs/89R/billtext/html/HB00149F.htm
- FTC, final rule banning fake reviews and testimonials (August 2024): https://www.ftc.gov/news-events/news/press-releases/2024/08/federal-trade-commission-announces-final-rule-banning-fake-reviews-testimonials
- EU AI Act, Article 50 (European Commission AI Act Service Desk): https://ai-act-service-desk.ec.europa.eu/en/ai-act/article-50
- European Commission, Quick Facts: Transparency rules for AI systems: https://digital-strategy.ec.europa.eu/en/factpages/quick-facts-transparency-rules-ai-systems
- Anthropic, Usage Policy (effective November 12, 2026): https://www.anthropic.com/aup
- CFPB, 12 CFR 1030.8, Advertising (Regulation DD): https://www.consumerfinance.gov/rules-policy/regulations/1030/8/