Is AI UGC legal? What the FTC, TikTok and the EU actually say

Yes in the US, when it’s honest: the FTC says its 2024 reviews rule has “no blanket prohibition” on AI avatars in marketing, but a made-up person can’t claim she used a product, and New York now requires ads with AI performers to say so (from June 2026). In the EU, the AI Act has added a duty to disclose AI deepfakes since 2 August 2026. TikTok Shop adds its own ban on AI “experts” pushing health products. AI UGC is a subject of its own, so this page goes deeper than our AI influencer rules guide. Checked 2 October 2026. Not legal advice.

On this page
  1. Is AI UGC legal in the US?
  2. Where is the line between AI UGC and a fake testimonial?
  3. Does a paid AI UGC post need an #ad?
  4. What does TikTok allow for AI UGC?
  5. Does New York require AI ads to be labelled?
  6. What does the EU AI Act require for AI UGC?
  7. What should your agreement with the brand say?
  8. What can go wrong?

Yes. The FTC’s Consumer Reviews and Testimonials Rule, in effect since 21 October 2024, doesn’t ban AI avatars. Its own Q&A says the rule “has no blanket prohibition on the use of AI-generated avatars in marketing” and was drafted “so as to not prohibit companies from using virtual influencers”. The problem is fake testimonials, whoever or whatever delivers them.

From the FTC’s rule Q&A and its announcement of the rule (14 August 2024):

  • An avatar used in a testimonial is prohibited “only if the underlying testimonials were fake or false”.
  • The rule targets testimonials that misrepresent that they’re “by someone who does not exist, such as AI-generated fake reviews”, or by someone without real experience of the product.
  • It bans writing or selling fake reviews and testimonials, and buying or spreading them when the business knew or should have known they were fake. Whoever makes the video is the one writing it.
  • Using a celebrity avatar without permission to praise a product violates the rule, “if reasonable consumers would think that the celebrity actually gave a testimonial”.
  • Even where the rule doesn’t apply, “the use of avatars could also be deceptive in violation of the FTC Act”, the FTC adds, pointing to its Endorsement Guides.
  • An “obviously fictional dramatization” isn’t a testimonial.
  • The FTC can now seek civil penalties against knowing violators.

Where is the line between AI UGC and a fake testimonial?

At experience. Your character can hold the product, show it and say what it is. She can’t say she used it, liked it for a month or got results, because she doesn’t exist and never did. Write every script around the product and what the brand can stand behind, never around her life.

Our reading of the FTC rule, applied to UGC scripts:

Script line OK? Why
“This is the new 500 ml bottle, it fits in a gym bag.” Yes A fact about the product, shown on screen
“Here’s how you use it: two pumps, then rinse.” Yes A demonstration of the product
“I’ve used this every day for a month and my skin cleared up.” No A made-up person claiming an experience and a result
“As a nurse, I recommend this.” No A fake credential on top of a fake person
A character who looks like a famous actress, praising the product No Celebrity likeness without permission

Does a paid AI UGC post need an #ad?

When the post runs on your character’s own account and a brand paid for it, yes. The FTC’s Endorsement Guides define an endorser as someone who “could be or appear to be an individual”, which covers an AI character. The FTC also says that for video, the disclosure belongs in the video itself, not only in the description or bio.

The sources: the Endorsement Guides, 16 CFR Part 255, and the FTC’s Disclosures 101 for social media influencers, which also warns: “Don’t assume that a platform’s disclosure tool is good enough.”

So a sponsored post by an AI character carries two separate disclosures: that she’s AI (the platform’s AI label) and that it’s paid (#ad or similar). One doesn’t replace the other. That’s our reading of two separate rules, not a single FTC line. Videos you hand to a brand to run as its own ads are the brand’s ads; the AI disclosure then has to travel with them, which is why it goes in your agreement.

What does TikTok allow for AI UGC?

TikTok has three layers, and AI UGC can hit all of them. The community guidelines require realistic AI content to be labelled. The ads policy allows AI content with a label or clear disclaimer. TikTok Shop’s seller rules go furthest: they ban AI “doctors” or “professors” promoting health products and AI that misrepresents how a product looks or works.

From TikTok Shop’s US seller AI-generated content rules (the page is dated 09/02/2026; read 2 October 2026):

  • Content with “synthetic faces, voices, digital humans, or highly realistic virtual figures” must be labelled as AI.
  • No fake professionals such as “doctors” or “professors” to promote health-related products.
  • AI “must not be used to demonstrate a core product effect that is not supported by the actual product”.
  • AI must not replace or fabricate product displays in a way that misleads about the actual product.
  • Disclose with on-screen text, a watermark, a sticker or a note, or turn on the “AI-generated content” toggle, which shows “Disclosed by creator as AI-generated”.
  • Penalties run from a warning up to permanently losing commission withdrawals or product promotion rights, or a suspended or banned account. Using AI content for fraud can mean a permanent ban.

On the rest of TikTok: realistic AI content must be labelled under the community guidelines effective 24 September 2026, and the ads policy allows AI-generated content with the AI label or your own clear disclaimer.

Does New York require AI ads to be labelled?

Yes. New York’s S8420-A, signed on 11 December 2025 as Chapter 617, requires a business that produces or creates an ad for a commercial purpose to disclose conspicuously that it contains a “synthetic performer”, when they know it does. It took effect 180 days after signing, in June 2026. Penalties: $1,000 for a first violation, $5,000 after.

A synthetic performer is an AI-made asset meant to look like a human performance that isn’t recognisable as any real performer. An original AI character in a video ad fits that description. Ads for films, TV and games, audio-only ads and AI translation are exempt (New York Senate, S8420-A, read 2 October 2026). Other US states may have their own rules; we haven’t checked them.

What does the EU AI Act require for AI UGC?

Since 2 August 2026, the EU AI Act requires people who publish AI deepfakes to disclose them. The duty sits with “deployers”, which excludes purely personal use, and making videos for paying brands is professional work, so it likely applies to you and to the brand. EU consumer-protection law applies on top of that; this page doesn’t cover it.

The dates, fines and the debate over whether an invented character counts as a deepfake are in our AI influencer rules guide. What the labelling duty asks of a creator, step by step, is in EU AI Act labelling requirements.

What should your agreement with the brand say?

That the video is AI-generated, in plain words, before the brand pays. The course’s rule for Phase 5 is simple: tell the brand it’s AI and put it in the agreement. Then add the few lines that keep both of you on the right side of the rules above. Our checklist is below; for a large contract, ask a lawyer.

  1. The video is AI-generated and features an original AI character, not a real person.
  2. The character doesn’t claim personal experience of the product or results.
  3. The brand supplies any product claims in the script and can back them up.
  4. The brand labels the ad as AI where a platform or law requires it (TikTok, TikTok Shop, New York and others).
  5. No fake credentials: she doesn’t play a doctor, nurse or other expert.
  6. No real-person likeness: the character doesn’t resemble a celebrity or anyone real.
  7. The product is shown as it really is: no AI changes to its size, colour or what it does.

The production side, including how to check the product in every clip, is in how to make AI UGC videos. What to charge for it is in AI UGC pricing.

What can go wrong?

  • She says “I use this every day”. That’s a fake testimonial under the FTC rule. Script the product, not her experience.
  • The brand adds a results claim later. Your video now makes a claim neither of you can back up. Agree that you approve script changes.
  • She’s dressed as a doctor for a supplement. TikTok Shop bans AI experts promoting health products, and repeat or serious violations can get the account suspended or banned.
  • The product looks better than it is. A generated bottle with a cleaner label or a bigger size misleads buyers. Regenerate until it matches.
  • The AI label gets dropped in editing. The brand reposts without it and the platform rejects or removes the ad. Put labelling in the agreement.
  • Her face resembles someone famous. A celebrity likeness praising a product without permission is a problem under the FTC rule and on every platform. Change the character, and see where AI likeness laws stand.
  • You rely on this page alone. Rules change. Check the linked sources, and get legal advice for anything serious.

Questions people ask

Is it legal to use an AI avatar in an ad?
In the US, yes. The FTC says its Consumer Reviews and Testimonials Rule has no blanket prohibition on AI-generated avatars in marketing. The avatar becomes a problem when it gives a fake or false testimonial.
Can my AI character say she uses the product?
No. She doesn't exist and never used it, so that's a testimonial that misrepresents the person's experience, which the FTC's rule bans. Have her show the product and say what it is, not what it did for her.
Do I need to label AI UGC on TikTok?
Yes. TikTok requires realistic AI content to be labelled, its ads policy allows AI content with a label or clear disclaimer, and TikTok Shop requires AI labels on content with synthetic faces or digital humans.
Who is responsible if an AI UGC ad breaks the rules?
Possibly both of you. The FTC's rule bans writing or selling fake reviews and testimonials outright, which covers whoever makes the video, and bans buying or spreading them when the business knew or should have known they were fake, which covers the brand. Keep the script honest and the AI disclosed in writing.

Read next