The company that shows the world most of its online ads wants you to know when a machine helped make them. Google said it will begin disclosing when an advertisement was created using artificial intelligence, surfacing the information in a "How this ad was made" panel reachable from the menu on ads across Search, YouTube and Discover, TechCrunch reported. It extends a disclosure Google previously applied only to election ads to advertising more broadly.
How it works
The mechanism has two tracks, Google said. When an advertiser uses Google's own generative-AI tools to build an ad, the disclosure is added automatically. When an advertiser uses outside AI tools, the onus is on them to flag the ad as AI-made; Google is not independently verifying every such claim. A user who wants to check taps the menu on an ad and reads the "How this ad was made" details.
Google also says it is leaning on a technical standard called C2PA "content credentials", a kind of tamper-resistant digital label embedded in an image or video that records how the file was made and whether AI was involved. Google already attaches these credentials to images produced by its own AI models, and the plan is to use them to make disclosures more reliable over time.
Why Google is doing it
Two forces are pushing this. The first is the sheer spread of AI-generated ad creative: it is now cheap and fast to conjure product images, voices and video with AI, and that has fueled worries about deception, deepfakes and manipulation, especially around elections. Regulators in the US and Europe have been moving toward requiring disclosure of synthetic media, and platforms are getting ahead of, or responding to, that pressure.
The second is that advertising is Google's lifeblood. Ads generate the large majority of parent company Alphabet's revenue, so how trustworthy those ads are is not a side issue for Google; it is central to the product. A flood of unlabeled, AI-faked ads would erode the trust the whole business rests on.
How it compares, and its limits
Google is not first here. Meta, which runs Facebook and Instagram, has for a couple of years required labels on ads made or significantly edited with AI and has leaned on automated detection, applying "Made with AI" labels that advertisers cannot simply remove. Against that, Google's reliance on advertisers to self-report when they use third-party tools looks lighter-touch.
That is the obvious weakness. A disclosure system that depends on honesty is only as good as advertisers' willingness to comply, and there is no automatic check on ads made with outside tools. Google can penalize violators by disapproving ads or issuing account strikes, but enforcement of a self-reported rule is inherently harder than catching something a system can detect on its own.
Why it matters
The move is small in mechanics but meaningful in signal. The dominant force in digital advertising is conceding that users have a right to know when what they are looking at was generated by AI, a principle that, once established for ads, is hard to contain. For the broader information economy, it is another step in a slow shift toward labeling synthetic content, driven by a mix of regulation, reputational risk and public unease. Whether it changes what people actually see, or becomes a box advertisers tick and ignore, will depend on how seriously it is enforced. This article is informational and not investment advice.
