Are AI UGC Ads Allowed on Meta and TikTok?

Short answer: yes. AI-generated UGC ads are allowed on Meta and TikTok, and brands run them at scale every day. What both platforms require is disclosure. If your ad contains photorealistic video or realistic-sounding audio that was created or altered with AI, you are expected to label it, and both platforms are increasingly able to detect and label it themselves if you do not.

The confusion is understandable. Headlines about AI content rules tend to blur three separate things: what the platforms actually require of advertisers, what they require of ordinary posters, and what a handful of state laws require in political advertising. This post separates them, using the platforms' own published policies as the source, so you can run AI UGC creative without waking up to a rejected ad or a strike on your account.

Are AI-generated ads against Meta's policies?

No. Meta does not prohibit AI-generated ad creative. Its policy is a disclosure policy, not a ban. Meta's own community standards state that people must disclose, using its AI-disclosure tool, whenever they post content containing photorealistic video or realistic-sounding audio that was digitally created or altered. Meta also applies its own informative labels to digitally created or altered content, and it reserves stronger labeling for content that carries a particularly high risk of materially deceiving the public on a matter of public importance.

Two practical things follow from that. First, the trigger is realism, not the use of AI in general. Using AI to color-correct, crop, upscale, or draft a headline is not what the policy is aimed at. A synthetic human who looks like a real person talking to camera is squarely what it is aimed at. Second, the enforcement risk is not "you used AI." It is "you used AI to make something that looks real, and you did not say so."

Does TikTok allow AI-generated content in ads?

Yes, with the same core condition. TikTok's community guidelines require creators to label AI-generated content that contains realistic images, audio, or video, and TikTok provides an AI-generated label for exactly this purpose. In its own announcement of the label, TikTok said creators may apply it to any content that has been completely generated or significantly edited by AI, and that other disclosures, such as a sticker or a caption, also satisfy the requirement.

TikTok has also said it is working toward applying an AI-generated label automatically to content it detects was created or edited with AI. That is the part advertisers should plan around. If the platform can label your ad for you, the choice is not really whether the ad is marked as AI. The choice is whether you marked it yourself, which looks deliberate, or the system marked it for you, which looks like something you were trying to avoid.

On top of the AI label, TikTok requires a separate commercial disclosure for paid and branded content. These are two different switches. An AI-generated ad promoting your own product needs both the commercial disclosure that comes with running it as an ad and the AI content label if the video is photorealistic.

What actually gets an AI UGC ad rejected

In practice, almost nothing gets rejected for the simple fact of being AI. The rejections come from things that would have been violations with a human creator too, and AI just makes them easier to commit by accident:

• Undisclosed synthetic humans. A photorealistic presenter who is not labeled, in a category where realism matters, is the single most common problem.
• Using someone's likeness or voice without consent. An AI presenter based on a real, identifiable person needs documented permission. Cloning a public figure, a celebrity, or a competitor's founder is the fastest route to a suspended account, and it is a legal problem, not just a policy one.
• Fabricated testimonials and claims. If a synthetic creator says "I lost 20 pounds in three weeks," you have made an unsubstantiated claim and put it in the mouth of a person who does not exist. Both platforms and the Federal Trade Commission care about that. The FTC's rule on fake reviews and testimonials does not have an exemption for AI.
• Restricted categories. Health, finance, and dating creative already sits under stricter review. Synthetic presenters making before-and-after or earnings claims in those categories is the highest-risk combination there is.

Read that list again and notice what it has in common. Every item is about deception, not about the technology. Platforms are relaxed about AI creative and unforgiving about ads that mislead people. Keep the claims true and the disclosure honest and you can run AI UGC all day.

Do I have to label an AI UGC ad?

If the ad contains a photorealistic person, or realistic-sounding voice, that was generated or significantly altered by AI, then yes: label it. That is the plain reading of both platforms' policies, and it is also the pragmatic call, because detection keeps improving and self-labeling is the version of this that does not cost you anything.

Here is what disclosure looks like on each platform in practice.

Both policies are living documents and both companies update them regularly. Treat the table above as a map, not as a substitute for the current policy page on the day you launch.

Does the AI label hurt ad performance?

This is the question advertisers actually care about, and the honest answer is that there is no credible public dataset that isolates the effect of the label itself. Anyone quoting you a precise percentage is guessing.

What can be said with more confidence is that the label is not the variable that decides whether an ad works. Paid social performance is dominated by the hook, the angle, and the offer. A well-cast creator saying something genuinely interesting in the first two seconds beats a poorly-conceived ad with no label on it, every time. The audience that scrolls past your ad in 1.5 seconds never got far enough to read a label anyway.

There is also a plausible upside that gets ignored. Viewers are getting better at spotting synthetic video on their own. An ad that looks AI-made and is not labeled reads as a small deception. The same ad with a label reads as a company being straightforward. Over time, the honest version is the one that keeps its audience.

How to run AI UGC ads without policy trouble

A workable checklist, in the order you should think about it:

1. Use synthetic creators, not synthetic people. Your presenter should be a generated person who resembles no one in particular. Never a real, identifiable individual without written consent.
2. Keep every claim substantiated. The script has to pass the same review your website copy would. If you cannot prove it, the creator cannot say it, no matter how well it converts.
3. Do not stage a fake testimonial. A generated creator describing a benefit is marketing. A generated creator pretending to be a verified customer sharing personal results is a fabricated testimonial. The line is real and worth staying well behind.
4. Label it, in the platform's own tool. Use the AI-disclosure tool on Meta and the AI-generated label on TikTok when the content is photorealistic. Do it at upload, every time, so it never depends on someone remembering.
5. Check the policy page each quarter. These rules have moved more in the last two years than in the previous ten. Put a recurring reminder on the calendar of whoever owns the ad account.

None of that is heavy lifting. It is fifteen minutes of process that lets you use the fastest creative pipeline available to you without ever wondering whether the account is at risk.

The competitive reality

While a lot of brands are still debating whether AI UGC is allowed, their competitors have quietly moved on and are testing eight new concepts a week with it. You do not have to take that on faith: the ad libraries are public, and it is worth an afternoon to look at the video creative your competitors are running right now and count how much of it is clearly synthetic. In most direct-response categories, the answer surprises people.

That is the real cost of the confusion. Not a policy strike, but the months spent waiting for clarity that already exists, while the accounts that read the actual policy and shipped are compounding creative learnings you do not have.

Where to go from here

If you are past the question of whether you are allowed to do this and into the question of how to do it well, the two things that matter most are the angle and the volume. Our breakdown of AI UGC ad examples by angle covers the concepts that consistently work, and how many ad creatives you should be testing covers how much of it you need to be shipping.

And when you want the ads themselves, that is what AdsGen does. Paste a product URL and get a batch of creator-style video ads with the angle chosen, the hook written, captions burned in, and native vertical sizing, ready to upload and label. If you are running paid social on Meta specifically, our Facebook ad generator page walks through what that looks like end to end.

This article summarizes publicly available platform policies as of July 2026 and is not legal advice. Meta and TikTok update their policies frequently. Always check the current policy in your ads manager before you launch.