Samsung warns TikTok’s AI‑ad policy is failing, urges stricter enforcement now.

Samsung’s internal compliance team has apparently flagged the discrepancy between its TikTok and YouTube ad pipelines, according to a report by The Verge. The company’s marketing assets for the Galaxy S26 Ultra’s “privacy display” were uploaded to YouTube with explicit AI‑generation disclosures embedded in the collapsed description fields, yet the same videos appeared on TikTok without any of the platform‑mandated AI labels. The Verge notes that “regular videos on Samsung’s TikTok accounts – those not actively promoted as ads – also lack AI disclosures, despite those same videos being labeled as AI‑generated on YouTube.” This suggests a systematic failure to propagate metadata from Samsung’s content‑management system to TikTok’s ad‑submission interface, rather than an isolated oversight.

TikTok’s advertising policy, as quoted by The Verge, requires any “significantly edited or generated by AI” content to carry a clear label, which can be a platform‑provided AI tag or a custom disclaimer, watermark, or sticker. The policy defines “significantly modified by AI” as alterations that go beyond minor tweaks, encompassing fully synthetic imagery, video, or audio, as well as substantial transformations of real source material. The Verge points out that the Samsung ads in question were “disclosed as an ad when it appeared on my timeline, but not that AI was used to make it,” indicating that the platform’s enforcement mechanisms either did not detect the AI‑generated metadata or failed to apply the required label during the ad review process.

Both Samsung and TikTok are signatories of the Content Authenticity Initiative (CAI), which promotes the adoption of the Coalition for Content Provenance and Authenticity (C2PA) standards for provenance metadata. The Verge emphasizes that “TikTok and Samsung supposedly share similar ideals regarding the labeling of AI content,” implying that the two firms have a mutual technical framework for embedding provenance tags. However, the report observes that “if Samsung knowingly used AI to make its videos, it should have told TikTok when the ads were submitted,” and conversely, “if TikTok was informed, it should have made sure its users were aware, per the platform’s own advertising policies.” The apparent breakdown suggests a gap between C2PA‑compliant metadata generation and the downstream enforcement tooling that TikTok employs for ad content.

In response to the findings, Samsung has reportedly issued an internal memorandum urging stricter adherence to TikTok’s labeling requirements and requesting that the platform tighten its verification pipeline. The Verge cites the company’s warning that “the AI‑ad policy is failing and demands stricter enforcement now,” framing the issue as a broader industry risk where advertisers and platforms that champion transparency may inadvertently undermine it through procedural lapses. Samsung’s push for tighter enforcement aligns with its broader AI‑driven marketing strategy, which relies on generative models to produce high‑volume visual assets; any ambiguity about the origin of those assets could erode consumer trust and expose the brand to regulatory scrutiny.

The Verge’s coverage does not provide quantitative data on how many Samsung videos are affected, but it does highlight a pattern: the same promotional material is treated differently across platforms despite identical source files. This inconsistency raises questions about the robustness of cross‑platform provenance pipelines, especially as more brands adopt generative AI at scale. If the C2PA metadata embedded in the video files is not being read reliably by TikTok’s ad‑review system, the industry may need to standardize not just the format of provenance tags but also the verification APIs that platforms must implement. Samsung’s call for “stricter enforcement” therefore signals a potential shift toward more rigorous, possibly automated, validation of AI‑generated content before it reaches consumers.