You built a killer ad with an AI-generated model, a synthetic voiceover, or a background your designer never touched with a camera. Someone on your team asks: “Do we need to label this?” You Google it, get three different answers, and now you’re more confused than when you started.
Here’s the uncomfortable truth: there’s no single answer. Meta, Google, and TikTok each wrote their own rulebook, and a disclosure that satisfies one platform can still get you flagged on another. Layer in a handful of new state laws, and “just add a disclaimer” stops being good enough advice.
We’re going to walk through what each platform actually requires, not what a LinkedIn post said it requires. Then we’ll show you how to build a disclosure process that doesn’t break every time a platform updates its policy.
Navigating divergent platform requirements demands a standardized internal framework. Brands cannot rely on inconsistent automated metadata detection to satisfy strict regulatory obligations. By implementing mandatory manual review protocols at the creative stage, digital advertisers ensure every campaign maintains platform compliance, avoids costly account suspensions, and adheres to evolving digital advertising disclosures.
Most brands think about AI disclosure the way they think about cookie banners: set it once, forget it. That’s not how this works anymore.
The mistake we see constantly is treating “AI disclosure” as one policy instead of four. Advertisers build a single internal rule usually “label anything AI-generated” and apply it the same way across Meta, Google, and TikTok. Then they wonder why an ad sails through review on one platform and gets rejected on another.
Each platform defines “AI-generated” differently, decides who applies the label differently, and penalizes a missed label differently. A color-corrected product photo might need nothing on TikTok and nothing on Meta. A fully synthetic spokesperson needs a label everywhere, and depending on where your audience lives, it might also need to satisfy a state law that has nothing to do with the platform at all.
Skip that distinction, and you’re not being cautious. You’re guessing and guessing is exactly what gets ad accounts flagged for misrepresentation
Here’s the part most advertisers miss: platforms didn’t converge on AI disclosure because they’re not solving the same problem.
Meta and Google built their disclosure systems around a single question: could this content mislead someone into thinking they’re seeing something real? That’s why both platforms auto-detect AI origin through embedded metadata and only strictly require self-disclosure in narrow categories like political and social-issue ads. For a standard product ad, Meta and Google are largely trusting you.
TikTok took the opposite approach. Its policy asks whether content could pass as real footage of a real person or event and if the answer is yes, disclosure isn’t optional, no matter what you’re advertising. According to TikTok’s advertising policy , significantly edited or AI-generated content is only allowed once you apply TikTok’s AIGC label or add your own clear disclaimer, caption, watermark, or sticker.
Then there’s a third layer that has nothing to do with the platform at all: the law. New York now requires advertisers to disclose “synthetic performers” in ads reaching New York residents, regardless of which platform the ad runs on. The New York Governor’s office confirmed the law took effect June 9, 2026, and it applies to the ad itself, not just the platform’s internal labeling system.
Platform policy and state law are two separate compliance problems. You can satisfy one and still violate the other.
Now here’s where it gets specific. Let’s break down exactly what each platform requires today.
Meta’s approach is built around a label called “AI Info,” shown inside the “About this ad” panel. Meta applies this label two ways.
For most standard product and service ads, that’s it Meta labels it for you. According to Meta’s Help Center, the harder requirement kicks in for ads about social issues, elections, or politics: those advertisers must disclose AI-created or AI-edited visuals themselves, on top of anything Meta detects automatically.
The catch: Meta’s detection depends on metadata that isn’t consistently read across every generative AI tool on the market. If your AI creative tool doesn’t embed the right signals, Meta may never catch it which sounds like a loophole until you remember TikTok and state law aren’t relying on the same detection system.
Google’s disclosure system, “How this ad was made,” rolled out gradually starting in July 2026 across Search, YouTube, Display, and Discover. Users open it from the three-dot menu on any ad.
The mechanism mirrors Meta’s:
Google is explicit that flipping this control on doesn’t guarantee legal compliance in every market. That responsibility sits with you, and it’s a good reminder that platform settings and legal disclosure are two different checkboxes. It’s worth pairing this review with a look at how the AI Max migration is already reshaping how your Search campaigns get built in the first place.
TikTok is the strictest of the three, and it’s the one advertisers underestimate most.
TikTok’s advertising policy draws a firm line between “significant” and “insignificant” AI edits. Insignificant edits color correction, cropping, brightness adjustments, background denoising need nothing. Significant AI use requires disclosure, full stop:
An AI-generated spokesperson holding your product isn’t a gray area. It’s “completely AI-generated” by TikTok’s own definition, and their policy explicitly tells advertisers: if you’re not sure, label it anyway.
There’s no metadata-detection safety net here the way there is with Meta and Google. TikTok gives you two ways to comply its built-in AIGC label in Ads Manager, or your own clear disclaimer, caption, watermark, or sticker but the burden to apply one of them sits entirely on you.
Even a perfectly labeled ad on Meta, Google, and TikTok can still violate a state or national law, because those laws don’t care which platform you used.
New York’s synthetic performer law requires a conspicuous, in-ad disclosure any time an advertisement includes a digitally created human that’s meant to look like a real, unidentifiable performer regardless of the platform. Penalties start at $1,000 for a first violation and climb from there.
California’s AI Transparency Act takes a different angle, requiring large AI providers to embed machine-readable and, where feasible, visible disclosure into content their tools generate. It became operative in August 2026, and it puts real obligations on the AI companies whose tools you’re using to build creative, not just on you.
The EU AI Act’s transparency provisions follow a similar logic for AI-generated or manipulated content shown to EU audiences.
None of these laws are optional just because your ad “already has an AI label” on the platform. Treat platform labeling and legal disclosure as two separate line items on your pre-launch checklist, not one.
Stop treating this asset by asset. Build it into your workflow once.
A photo shot on a real product, color-corrected in Photoshop, needs nothing. A photo generated from a text prompt needs a label on every platform where it’s realistic-looking.
Meta and Google auto-label their own AI tools. Third-party tools and TikTok require you to flag it manually, every time.
If the ad could reach New York or the EU and includes an AI-generated human, add an in-ad disclosure regardless of what the platform’s own label says.
Do this at the creative brief stage, not during ad review. Catching a missing disclosure after your media buyer has already scheduled the launch costs you time you don’t get back.
This is the exact audit we build into every AI-assisted campaign before a client’s ad account goes live because a rejected ad costs you a launch date, and an undisclosed synthetic performer in the wrong state costs you a lot more than that. If your team is already running Meta, Instagram, or TikTok ad campaigns Google Search and Display with AI-generated assets, this is worth getting right before your next launch, not after a rejection.
Stop guessing whether your ad needs a label. Let’s map your creative workflow against every platform’s actual policy book a free audit call at www.themayk.com
Navigating AI ad disclosures requires balancing platform policies with strict state laws. Because Meta, Google, and TikTok enforce different rules for AI-generated creative, advertisers cannot rely on a single approach. To avoid account rejections and legal penalties, teams must review assets early, manually flag synthetic content, and ensure compliance before launching campaigns.
SEO – unlock sustainable growth with proven search strategies.
Content Strategy – magnetic content that earns links, shares, and brand authority.
Paid Media – precision campaigns built for measurable ROI.
Do You Have to Label AI-Generated Ads on Meta, Google And Tik Tok? You built a killer ad with an AI-generated model, a synthetic voiceover, or a background your designer never touched with a camera. Someone on your team asks: “Do we need to label this?” You Google it, get three different answers, and now […]
Why Is My Google Merchant Center Account Suspended for Misrepresentation? Your Merchant Center account got flagged for misrepresentation, and there was no warning email, no grace period, nothing. One day your Shopping ads were live. The next, every product was gone. Support keeps pointing you back to the misrepresentation policy page, and it reads like […]
Google Ads AI Max Migration: What Changes on September 1? On September 1, Google starts automatically upgrading Search campaigns to AI Max. No click required from you. Google’s own messaging says the transition will “mirror” what you’re already running. That’s true for the interface. It’s not fully true for how your ads get triggered. If […]
How we use your info: Used strictly to review your project, prepare tailored insights, and reach out for a 1-on-1 strategy call. Zero spam, zero shared data.