Two FTC texts govern an AI UGC ad in the US: the Endorsement Guides (16 CFR Part 255, revised June 29, 2023) and the Trade Regulation Rule on the Use of Consumer Reviews and Testimonials (16 CFR Part 465), which the FTC says "went into effect on October 21, 2024." Neither bans AI creators. The FTC's own Q&A says the reviews rule "has no blanket prohibition on the use of AI-generated avatars in marketing." What both forbid is an AI person presented as a real customer with a real experience that did not happen. This page quotes the source texts as of 2026-11-09 and maps them to choices you make on an 8frame board. It is general information, not legal advice.
For platform AI labels and state laws on synthetic performers, see do you have to disclose AI-generated ads. This page is only about the testimonial question: who is speaking, and are they who the ad says they are.
The two texts, in their own words
The reviews rule (16 CFR 465), in effect since October 21, 2024
Section 465.2(a) makes it unfair or deceptive for a business to write, create or sell a consumer review, consumer testimonial or celebrity testimonial that materially misrepresents, expressly or by implication:
- "(1) That the reviewer or testimonialist exists;"
- "(2) That the reviewer or testimonialist used or otherwise had experience with the product, service, or business"
- "(3) The reviewer's or testimonialist's experience with the product, service, or business"
The FTC's announcement of the final rule (August 14, 2024) names the AI case directly: the rule covers "reviews and testimonials that misrepresent that they are by someone who does not exist, such as AI-generated fake reviews", and it lets the FTC "seek civil penalties against knowing violators."
The Endorsement Guides (16 CFR 255), revised June 29, 2023
- 255.1(a): "Endorsements must reflect the honest opinions, findings, beliefs, or experience of the endorser."
- 255.1(c): when the ad says the endorser uses the product, "the endorser must have been a bona fide user of it at the time the endorsement was given."
- 255.1(g): "The use of an endorsement with the image or likeness of a person other than the actual endorser is deceptive" when it misrepresents a material attribute of the actual endorser.
- 255.2(b): an ad relating a consumer's experience will likely be read as representing "what consumers will generally achieve with the advertised product", and if that is not substantiated the ad "should clearly and conspicuously disclose the generally expected performance in the depicted circumstances."
- 255.2(c): ads presenting endorsements by what are represented to be "actual consumers" "should utilize actual consumers in both the audio and video," or "clearly and conspicuously disclose that the persons in such advertisements" "are not actual consumers of the advertised product."
The FTC's June 29, 2023 release says the revision changed the definition of endorsements to "clarify the extent to which it includes fake reviews, virtual influencers" and social media tags.
What "clear and conspicuous" means
Section 255.0 defines it: a disclosure that is "difficult to miss (i.e., easily noticeable)" and "easily understandable by ordinary consumers." The FTC's endorsement Q&A adds two practical points: "if the endorsement is made through visual means, the disclosure should be made at least visually", and a platform's built-in disclosure tool is "no guarantee that it's an effective way" to disclose.
What the FTC's Q&A says about AI avatars and actors
The FTC's Q&A on the reviews rule answers the two questions AI UGC raises:
- AI stock avatars. No blanket ban. An AI-generated "stock avatar" is not giving a "consumer review" as the rule defines it. Its use can still be a testimonial, which the rule prohibits only if the underlying testimonial is fake or false, and it can still be deceptive under the FTC Act with reference to 255.1(g) and 255.2(c).
- Actors portraying testimonialists. Not covered by the rule "unless the underlying testimonials were fake or false", but it "could be deceptive in violation of the FTC Act." In an obviously fictional dramatization, the actors are not giving testimonials.
- Celebrity avatars. Using a celebrity's avatar to praise a product without permission violates the rule if reasonable consumers would believe the celebrity endorsed it.
An AI creator is legally closer to an actor than to a customer. The question is always what the script claims.
The rules mapped to the board
| What the AI creator does in the ad | Which text applies | What to do on the board |
|---|---|---|
| Says "I've used this for a month and my skin cleared" | 465.2(a)(1) to (3): the testimonialist does not exist and did not use the product | Do not generate it. Rewrite as a presenter line ("This serum has niacinamide and..."), not a first-person experience |
| Presents features, holds the product, demonstrates use | not a testimonial; platform AI rules apply | Fine. Keep the claims to what the product does; label the ad as AI where the platform asks |
| Reads a real customer's review, with the customer's permission | 255.2(c): not an actual consumer on screen; 255.1(b) if presented as a quote | Disclose on screen that the person shown is not the customer, keep the customer's words intact, keep the permission on file |
| Uses a real person's face (yours, a colleague's, a creator's) | 255.1(g) likeness; consent and likeness law | Written consent for ad use. Never a public figure's face |
| Shows a "result" (before and after, a transformed room, a fitted body) | 255.2(b) typical results; TikTok bans misleading before-and-after | Show only what is substantiated and typical; on TikTok avoid product effect comparisons entirely |
| Is a paid human creator in a whitelisted ad | 255.5 material connection | Disclose the paid relationship in the video itself; see UGC whitelisting and Spark Ads |
On 8frame, the place these choices happen is the script and the prompt, not a setting. There is no "disclosure" node. You add the disclosure as a text or caption layer in the Video Editor, which also has caption layers generated from a pasted script, so put the disclosure where it is "difficult to miss": on screen, not only in the caption of the post.
Writing the script so it stays on the right side
The UGC hook formulas page has openers. Some of them are first-person by design ("I was today years old when..."). With an AI creator, check every line against three questions:
- Does the line claim the speaker used the product? If yes and the speaker is generated, cut it or attribute it ("Our customers tell us...", with a real source).
- Does it describe a result? If yes, is it typical, and could you show the evidence?
- Would a viewer think this is a real customer? If yes, either make it obviously a presenter or disclose that the person is not a customer.
Hook lines that survive all three: questions ("Ever opened a bottle that leaked in your bag?"), demonstrations ("Watch what happens when I drop it"), and product facts ("This lid has a second seal"). They also suit AI production: a demonstration beat is image-to-video from a product still, which is the part of the board AI handles best.
The platform layer on top
The FTC rules are about deception. Platforms add their own labels:
- TikTok's ad policy allows AI-generated content with its AIGC label or your own clear disclaimer, rejects or restricts undisclosed AI content, and bans misleading before-and-after product comparisons.
- Meta labels ads made or significantly edited with its own generative tools with "AI info", places the label next to "Sponsored" when its tools add a photorealistic AI human, and says it will also begin detecting ads made with third-party AI tools. Both platforms' rules are collected on do you have to disclose AI-generated ads.
A platform label does not fix a fake testimonial. An ad can carry the AIGC label and still break 465.2 if the AI person claims an experience nobody had.
Checklist before you export
- Every first-person experience line traces to a real customer who gave permission.
- No generated person is presented as a customer without an on-screen disclosure.
- No real person's face or voice without written consent; no public figures at all.
- Results shown are typical and substantiated, or disclosed.
- Disclosure is visual if the claim is visual, and audible if the claim is spoken.
- Platform AI label switched on in Ads Manager.
FAQ
Is it legal to use an AI avatar in a UGC ad in the US? The FTC says its reviews rule "has no blanket prohibition on the use of AI-generated avatars in marketing." What is prohibited is a fake or false testimonial, whoever delivers it, and the FTC Act still applies to deceptive use.
Can an AI creator say "I love this product"? Not as a customer testimonial. A generated person did not use the product, which is what 465.2(a)(2) covers. Present them as a presenter, or use a real customer's words with permission and disclose that the person on screen is not that customer.
When did the fake reviews rule take effect? October 21, 2024, according to the FTC's Q&A. The FTC announced the final rule on August 14, 2024.
Is a disclosure in the post caption enough? The FTC says a visual claim needs at least a visual disclosure and that platform disclosure tools are "no guarantee" of an effective disclosure. Put it in the video.
Does the FTC require an "AI-generated" label? The texts quoted here are about deception, not about labelling AI as such. Labels come from platforms (TikTok AIGC, Meta "AI info") and some state laws. See the disclosure page linked above.
Sources
- 8frame prices, defaults and limits: the 8frame app, checked 2026-11-09.
- 16 CFR 465.2 (Cornell LII): https://www.law.cornell.edu/cfr/text/16/465.2, checked 2026-11-09.
- FTC press release, final rule banning fake reviews and testimonials (2024-08-14): https://www.ftc.gov/news-events/news/press-releases/2024/08/federal-trade-commission-announces-final-rule-banning-fake-reviews-testimonials, checked 2026-11-09.
- FTC, Consumer Reviews and Testimonials Rule: Questions and Answers: https://www.ftc.gov/business-guidance/resources/consumer-reviews-testimonials-rule-questions-answers, checked 2026-11-09.
- 16 CFR 255.0, 255.1, 255.2, 255.5 (Cornell LII): https://www.law.cornell.edu/cfr/text/16/255.0, https://www.law.cornell.edu/cfr/text/16/255.1, https://www.law.cornell.edu/cfr/text/16/255.2, https://www.law.cornell.edu/cfr/text/16/255.5, checked 2026-11-09.
- FTC press release, updated Endorsement Guides (2023-06-29): https://www.ftc.gov/news-events/news/press-releases/2023/06/federal-trade-commission-announces-updated-advertising-guides-combat-deceptive-reviews-endorsements, checked 2026-11-09.
- FTC, Endorsement Guides: What People Are Asking: https://www.ftc.gov/business-guidance/resources/ftcs-endorsement-guides-what-people-are-asking, checked 2026-11-09.
- TikTok Ads Policy, Misleading and false content (last updated April 2026): https://ads.tiktok.com/help/article/tiktok-ads-policy-misleading-and-false-content, checked 2026-11-09.
- Meta, "Expanding GenAI Transparency for Meta's Ads Products" (updated 2026-06-01): https://about.fb.com/news/2025/02/gen-ai-transparency-metas-ads-products/, checked 2026-11-09.
Try it without signing up: the 8frame canvas opens on a real board. The canvas is free and unlimited; generation is paid from $19/month.