Google is giving people a way to check whether the ad in front of them was built by a machine. The company announced on July 9 that it is rolling out transparency tools that reveal when generative AI has been used to create or edit an advertisement, a shift that touches billions of ads across its biggest platforms.
The centerpiece is a section called "How this ad was made," which now sits inside the My Ad Center panel on Google Search, YouTube, and Google Discover. Anyone, anywhere in the world, can open it by tapping the three-dot menu or the info icon on an ad. Until this week, that panel mostly told users who paid for an ad and why they were seeing it. Now it also tells them how the ad was put together.
What is actually changing
The move extends AI disclosure from a narrow corner of Google's ad business to nearly all of it. For years the company only demanded that election advertisers reveal synthetic or digitally altered content, a policy it introduced back in 2023. Commercial advertisers faced no such rule. That gap is what the new feature closes.
Keerat Sharma, Google's VP and General Manager of Ads Privacy and Safety, framed the update as a response to how quickly AI has moved into the advertising workflow. "As generative AI opens up new ways for businesses large and small to create ads, we're introducing additional transparency features across our advertising products," Sharma wrote in the company's announcement. The goal, he added, is to "help people better understand the ads they see, while providing advertisers with straightforward tools to navigate evolving industry standards."
The panel itself is simple. Open it, and it indicates whether an ad was created or edited with AI. What sits behind that simplicity is a two-track system that determines whether the disclosure shows up at all.
Two very different paths to a label
The reach of this feature depends heavily on which tools an advertiser used to build the ad.
When a brand uses Google's own generative AI advertising tools, the disclosure is switched on automatically. "We want to make managing AI disclosures as simple as possible for advertisers," Sharma explained. "So when they use Google's generative AI advertising tools to create ads, we'll automatically add a disclosure to each ad's My Ad Center panel."
The second path is where things get looser. If an advertiser builds an ad using a third-party tool, say a separate image generator or a rival voiceover service, Google hands them a control to flag that AI was involved. Crucially, the company has said it will not run its own check to verify whether that claim is true. The label rests entirely on the advertiser choosing to be honest. "When they create ads elsewhere, we're introducing a control so they can easily indicate if they used generative AI," Sharma wrote.
In some markets, the disclosure will not stay tucked inside the panel. Depending on local law, a label may appear directly on the ad itself, either automatically or once an advertiser flips the control, a way to keep advertisers compliant across jurisdictions without forcing a single global standard.
The honesty problem at the center
That third-party gap has drawn the sharpest scrutiny, and for good reason. An advertiser hoping a synthetic scene will pass for a genuine product photo has little incentive to volunteer otherwise, and Google is not looking over anyone's shoulder. The company controls its own AI systems, so labeling those outputs is trivial. For everything generated outside its walls, enforcement leans on advertiser disclosures and whatever provenance data travels with the file.
The imbalance is baked into the design. Inside Google's ecosystem the disclosure is automatic and dependable; anywhere else it is opt-in and unverified, so the honesty of the label tracks the honesty of the advertiser. A brand determined to hide its use of AI has plenty of room to do so.
Why Google is doing this now
The timing is not accidental. Generative AI has made it cheap and fast to produce slick product imagery and skip the cost of real-world e-commerce photography, a convenience for small businesses that is also a way to mislead a shopper who assumes they are looking at a real photograph. Google already prohibits misleading and deceptive ads, whether or not AI was used, and says those policies remain in force. "We continue to prohibit misleading and deceptive ads, whether created with AI or not, to keep the platform safe for everyone," Sharma wrote.
There is also a regulatory clock ticking. The European Union's AI Act carries transparency obligations for AI-generated content that are set to bite in August, and building disclosure tools ahead of the mandate positions Google to shape what compliance looks like in practice. Industry groups are already pushing back, with some retailers lobbying to exempt routine AI-assisted commercial ads from the strictest EU rules. In the UK, the Advertising Standards Authority's May 2025 guidance already requires disclosure when AI could mislead consumers about authenticity, and Google's tool makes that compliance easier to hit.
What it means for advertisers
For the brands running the ads, the update lands as both a convenience and a fresh obligation. Those already using Google's generative tools get compliance for free, since the platform handles the disclosure automatically. Teams that lean on outside AI software now carry the responsibility of flagging it themselves, and getting that wrong in a regulated market could invite trouble a single toggle would have avoided.
Advertisers operating across multiple regions will need to track where a label is mandatory, where it is optional, and where an on-ad overlay might appear. The safer posture, for brands that care about trust as much as reach, is to disclose by default rather than gamble on which jurisdiction is watching. Google is also not alone here. Meta now automatically labels ads created with its own AI tools and requires disclosures for third-party AI-generated ads across Facebook and Instagram, a sign the industry's direction of travel is set.
Building on older safeguards
Google presented the feature as an extension of transparency work already underway. The company embeds imperceptible signals, including its SynthID watermarking technology, into content produced by its generative AI tools, and it runs an advertiser verification program to confirm who is behind a given campaign. The new "How this ad was made" section adds context about the creative process to that existing scaffolding.
For consumers, the practical takeaway is modest but real. Click through an ad on Search or YouTube, open My Ad Center, and a disclosure will be waiting if one exists, though whether it exists comes down to which tools the advertiser used and whether they chose to be upfront about it.
That is the honest limit of what Google has shipped. The principle behind it, that people deserve to know when what they are looking at was generated or altered by AI, is hard to dispute. The enforcement mechanism is a different matter. For ads built with Google's tools, the label is automatic and dependable. For everything else, the company has built the disclosure and handed advertisers the switch. The honest ones will flip it. The rest are precisely why the label was needed in the first place.
Comments