TikTok · Policy · Compliance · Paid Social

TikTok AI Content Disclosure Rules for Advertisers (2026)

· 9 min read

Quick answer:if your TikTok ad contains AI-generated or significantly AI-edited footage, you must disclose it, either with TikTok's AIGC label or with your own clear caption, watermark, or sticker. TikTok's advertising policy states plainly that undisclosed AI-generated content will get the ad rejected or restricted. Separately, because the video promotes a product, you also need the Commercial Content Disclosure setting switched on. Those are two different toggles solving two different problems, and missing either one costs you distribution.

The three documents that actually bind you

Most of what gets written about TikTok and AI cites the Community Guidelines, which govern organic posts. If you are buying media, that is the wrong document to read first. Three pages matter, and they say slightly different things:

DocumentCoversLast updated
TikTok Advertising Policies, "Misleading and false content"Paid ads. Contains the section "Edited media and AI-generated content (AIGC)"April 2026
Community Guidelines, "Edited Media and AI-Generated Content (AIGC)"Everything posted, including the organic video behind a Spark Ad2025 H2 update, effective 13 September 2025
"About the Commercial Content Disclosure setting for advertisers"Brand/product promotion disclosure, unrelated to how the video was madeSeptember 2025

Worth noting what is notthere: TikTok's "Key TikTok Ad Policy Updates" changelog, current to August 2026, lists twelve policy changes in 2026 covering cosmetic surgery, alcohol, financial services, political advertising and more. None of them is an AIGC change. The AI rules for advertisers have been stable since the April 2026 refresh of the misleading-content policy.

What counts as AI-generated enough to need a label?

TikTok's advertising policy draws the line at "significantly modified by AI", which it defines as content changed beyond minor tweaks or enhancements. Here is the split, taken directly from that policy:

Needs a label (significant)No label needed (insignificant)
Images, video, or audio that are completely AI-generatedAdjustments to lighting, brightness, or colour saturation
The primary subject doing something they did not actually doRemoving or modifying backgrounds
The primary subject saying something they did not say, via AI voice-cloningDenoising images to reduce noise levels

For a typical AI UGC ad, this is not a close call. A synthetic presenter holding your product is "completely AI-generated" on the first row. Label it. TikTok adds its own tie-breaker: if you are uncertain whether your content falls under these categories, use caution and label it.

The Community Guidelines carve out three things that do not need disclosure on the organic side: small edits like colour correction, reframing or cropping; artistic styles such as anime; and generic text-to-speech narration, provided the TTS voice is not a recognisable voice of a known individual. That last one is the useful one. A stock synthetic voiceover over real product footage is treated differently from a cloned voice.

Where the label goes, and who applies it

You have three routes, and they are not interchangeable in effort:

  • Your own disclosure.TikTok's ad policy explicitly accepts a clear disclaimer, caption, watermark, or sticker that you burn in or write yourself. For ads uploaded straight into Ads Manager, this is usually the practical route.
  • The creator AIGC label.Applied in the app before posting. For a video it lives under More options after you tap Next; for a photo it is in Post settings. The label reads "creator labelled as AI-generated". TikTok warns that misleadingly applying it to unaltered content violates its Terms of Service.
  • TikTok's auto label.TikTok reads C2PA Content Credentials, provenance metadata that some AI tools attach, and auto-labels when it finds them. It also auto-labels anything made with TikTok's own AI effects. Once an auto label is applied, TikTok says you cannot remove it.

The auto label is the one that trips people up. Whether your ad arrives pre-labelled depends on whether your generation tool wrote Content Credentials into the file and whether your editing pipeline preserved them. Re-encoding a clip commonly strips metadata. Treat auto-detection as a backstop that may or may not fire, never as your compliance plan.

The second label everyone forgets

The Commercial Content Disclosure setting is a separate obligation. TikTok requires it on any post promoting a brand, product, or service, and it flags content as commercial on three signals: financial incentives (promo codes, URLs, collaboration mentions, QR codes), brand mentions (a hashtag, an @ tag, or a visible logo), and product recommendations (demos, tutorials, or a "buy now" call to action).

Enforcement here is faster than most advertisers expect. TikTok sends an in-app notification when a video looks like undisclosed branded content, and states these should be responded to within 24 hours before the video becomes ineligible for the For You feed and sees limited traffic distribution. Only the original poster can appeal, which is a real problem when a creator posted the video and your agency is managing the spend.

One useful exception: advertisers working with creators through TikTok One have the disclosure managed automatically, and TikTok states all TikTok One and TikTok Creative Exchange project videos posted to your Business Account are considered compliant.

Will labelling my ad as AI hurt performance?

TikTok says no, and it says it in two different places with two different levels of evidence, which is worth separating.

On the AIGC label, TikTok's support article states that turning on the AI-generated content setting will not affect the distribution of your video as long as it does not violate the Community Guidelines. That is an assertion, not a published study.

On the Commercial Content Disclosure setting, TikTok cites an actual test: nearly 2 million videos compared with and without proper commercial disclosure, with no performance difference reported (2023 TikTok Marketing Science, tested in Indonesia and Pakistan). Do not transfer that finding to the AI label. It is a different label measured in a different study in two specific markets three years ago. TikTok has not, as far as its public documentation shows, published equivalent numbers for AIGC.

What is documented is the downside of not labelling: ad rejection or restriction on the paid side, and removal, restriction, or forced labelling on the organic side.

What is still ambiguous

Anyone telling you TikTok's AI rules are fully settled has not read them side by side. Four genuine gaps:

  • "Encourage" versus "require".TikTok's support article says it encourages creators to label content completely generated or significantly edited by AI, then two sentences later says it requires labels on all AI-generated content containing realistic images, audio and video. The Community Guidelines and the ad policy both use the harder word. Assume required.
  • Fully synthetic people who are not anyone.The Community Guidelines' "disclosure is needed" examples are all framed around confusion about real people: face swaps, words someone did not say, cloned voices. An invented photorealistic presenter is not on that list, but it is squarely inside the ad policy's "completely AI-generated" row. The ad policy governs your ad, so label it.
  • Spark Ads and inherited labels. A Spark Ad runs an organic post, so both rule sets apply at once, but TikTok does not publish explicit guidance on whether an organic AIGC label satisfies the ad-side disclosure requirement. Belt and braces: label the organic post and keep a visible on-screen disclosure in the creative.
  • Local law sits on top.TikTok's own disclosure article notes that exact legal requirements vary by country and points US advertisers to the FTC Endorsement Guides under Section 5 of the FTC Act. Platform compliance is not legal compliance, and an AI presenter delivering a testimonial raises questions the platform policy does not answer.

Practical checklist before you upload

  • Is any part of the footage, voice, or scene AI-generated or AI-altered beyond colour, crop and denoise? If yes, disclose.
  • Add a visible on-screen disclosure (caption, sticker, or burned-in watermark) to the creative itself. It travels with the file and survives re-uploads.
  • If the video is posted organically first, switch on the AI-generated content setting before posting.
  • Switch on Commercial Content Disclosure for anything with a brand, product, promo code, or shop link.
  • Never use an AI likeness of a real person, public or private, without permission. TikTok prohibits false endorsements outright, labelled or not.
  • Watch for the in-app branded-content notification and respond inside 24 hours, and make sure whoever owns the posting account can appeal.
  • Check whether your generator writes C2PA Content Credentials. If it does, expect an auto label you cannot remove.
  • Keep a record of which ads used AI and how. When a rejection lands, you want to answer in minutes.

Sources

  • TikTok Business Help Centre, Advertising Policies, "Misleading and false content" (last updated April 2026)
  • TikTok Community Guidelines, Integrity and Authenticity, "Edited Media and AI-Generated Content (AIGC)" (2025 H2 update, effective 13 September 2025)
  • TikTok Support, "About AI-generated content"
  • TikTok Business Help Centre, "About the Commercial Content Disclosure setting for advertisers" (last updated September 2025)
  • TikTok Business Help Centre, "Key TikTok Ad Policy Updates" (last updated August 2026)
  • TikTok Newsroom, "Partnering with our industry to advance AI transparency and literacy" (C2PA Content Credentials)

All six were read directly in August 2026. Policies change; the dates above are your check for whether this page has gone stale.


Running the same creative on Facebook and Instagram? The rules are almost the opposite there: Meta's AI-generated creative ad policy puts the labelling burden on the platform, not you. On the creative side, how to make TikTok Shop ads with AI covers the build, and what makes AI UGC look fake covers the part policy cannot fix. UGC Vids AI generates the videos themselves, from $49/month, free for 3 days.

Frequently asked questions

Do I have to label AI-generated TikTok ads in 2026?

Yes. TikTok's advertising policy on misleading and false content (last updated April 2026) says significantly edited media and AI-generated content is allowed in ads only if you apply the AIGC label or add a clear disclaimer, caption, watermark, or sticker of your own. The same policy states that if TikTok identifies AI-generated content that has not been disclosed, the ad will be rejected or restricted.

What counts as 'significantly modified by AI' on TikTok?

TikTok defines it as content changed by AI beyond minor tweaks or enhancements. Its advertising policy gives three examples: content where the images, video, or audio are completely AI-generated; showing the primary subject doing something they did not actually do; and making the primary subject say something they did not actually say using AI voice-cloning. Minor tweaks that do not require a label include adjusting lighting, brightness, or colour saturation, removing or modifying backgrounds, and denoising.

Does the AI label reduce reach on TikTok?

TikTok's own support article states that turning on the AI-generated content setting will not affect the distribution of your video as long as it does not violate the Community Guidelines. TikTok has not published a performance study specific to the AIGC label. It has published one for the separate Commercial Content Disclosure setting, reporting no performance difference across nearly 2 million videos tested in Indonesia and Pakistan in 2023.

What happens if you do not label AI content on TikTok?

For paid ads, TikTok's advertising policy says undisclosed AI-generated content will be rejected or restricted. For organic posts, the Community Guidelines say unlabelled content may be removed, restricted, or labelled by TikTok's team depending on the harm it could cause, and realistic-looking content not yet confirmed as AIGC that presents matters of public importance in a potentially misleading way is ineligible for the For You feed.

Do I need two labels on a TikTok ad, one for AI and one for the brand?

Often yes, and they are separate settings. The AIGC label covers how the video was made. The Commercial Content Disclosure setting covers the fact that the post promotes a brand, product, or service, and TikTok requires it for any such post. TikTok flags commercial content based on financial incentives, brand mentions, and product recommendations, and undisclosed commercial content may lose For You feed eligibility.

Does TikTok automatically detect AI-generated videos?

Partly. TikTok reads C2PA Content Credentials, metadata attached by some AI tools, and may auto-apply an AI-generated label when it finds them. It also auto-labels content made with TikTok's own AI effects. TikTok notes that once an auto label is applied you cannot remove it from your post. Content Credentials are not attached by every generator, so auto-detection does not replace your own labelling duty.

Definitions

What is Spark Ads?What is AI UGC?What is TikTok Shop?What is Voice Cloning?What is UGC Creator?

Compare alternatives

UGC Vids AI vs ArcadsUGC Vids AI vs MakeUGCUGC Vids AI vs Creatify

Stop reading. Start shipping.

Generate your first UGC ad in 2 minutes. No editing required.

Try the free generator