Google has started showing users whether the ad in front of them was created or edited with AI. In the European Union, that disclosure now carries a legal duty behind it. Casino operators and B2B suppliers running Google Ads are inside the scope from day one.
The control sits in My Ad Center, the panel any user can open globally by clicking the three-dot menu or the info icon on an ad in Google Search, YouTube and Discover. The panel already lets people block an ad, report it, and see why it was shown. It now carries a line reading “How this ad was made”, which states whether AI was used to build the creative. Google announced the rollout on 9 July 2026.
“We want to help people better understand the ads they see, while providing advertisers with straightforward tools to navigate evolving industry standards,” said Keerat Sharma, vice president and general manager for ads privacy and safety at Google.
The disclosure is set two different ways. When an advertiser builds the ad with Google’s own generative advertising tools, the label is applied automatically. When the ad is made anywhere else, the advertiser has to set a new control declaring that AI was involved. Google does not run its own check to verify either answer. Google’s help documentation states that mandatory disclosure applies in the EU, India and New York, and that some AI use, such as cropping and standard editing, does not require a label. Until this change, election ads were the only category Google required to disclose synthetic or digitally altered content.
Where it bites in iGaming
Gambling ads sell on trust before they sell on price. A player comparing three casinos in a regulated market is already checking licence badges, payout terms and whether the brand looks legitimate. An AI notice attached to a bonus ad gives that player one more reason to hesitate, in a vertical where hesitation is the whole conversion problem.
The immediate click-through effect is likely to be small, because the notice is one click deep inside a panel most users never open. The same panel carries the report button, so a complainant now has a labelled starting point: an ad showing AI-generated slot art, a synthetic “winner” reaction shot or an invented game screen, sitting one click from a report form. Screenshots of that pairing travel well on social media and into regulator inboxes.
A slot thumbnail that does not match the actual game, or a mocked-up sportsbook interface built from prompts rather than the live product, also becomes harder to defend once the platform itself has recorded the creative as synthetic.
The EU turned it into compliance
Article 50 of the EU AI Act, which covers transparency obligations for AI-generated and manipulated content, applied from 2 August 2026. It requires deployers of AI systems that generate or manipulate image, audio or video content to disclose that the content is artificially generated, with providers required to mark outputs in a machine-readable format. Operators and suppliers advertising into EU markets carry that obligation whether or not Google asks them to tick a box.
National gambling advertising rules sit on top of it. The UK’s Committee of Advertising Practice (CAP) Code prohibits gambling ads that mislead about the chances of winning or the nature of an offer, and the Advertising Standards Authority (ASA) has ruled repeatedly on how games and prizes are depicted. Spain restricts most gambling advertising to a window between 1am and 5am under Royal Decree 958/2020. Italy has banned gambling advertising and sponsorship outright since the 2019 Dignity Decree. Poland’s competition authority has been probing seven licensed bookmakers over dark patterns, a reminder that presentation is now an enforcement subject in its own right.
The switch sits with affiliates and agencies
A large share of iGaming paid search is not run by operators directly. Affiliates, media buyers and performance agencies build and traffic the creative, so the self-declaration control lands with them, while the licensed operator carries the regulatory consequence and the brand damage.
Most affiliate agreements and creative briefs say nothing about AI-generated assets. Operators that do not add a clause are relying on a third party to make a compliance declaration on their licence.
Cheap creative just acquired a cost
Generative tools removed most of the expense of producing localised banner variants, game art and lifestyle imagery, which is why AI creative spread fast across casino acquisition campaigns and B2B supplier ads alike. The label does not stop any of that. It adds a disclosure decision to every asset, and a record that can be pulled up later.
B2B suppliers advertising platforms, aggregation and studio content to operator buyers face lower stakes. The audience is professional and the claims are commercial rather than consumer-facing. B2C acquisition teams are in a different position.
What marketing teams should check
- Audit live Google Ads creative and identify every asset built or edited with generative tools, including images sourced from suppliers and affiliates.
- Name an owner for the disclosure control in the account and write it into the workflow, so it does not fall to whoever uploads the asset.
- Add an AI-asset disclosure clause to affiliate and agency contracts.
- Check EU campaigns against Article 50 separately from Google’s control, since the legal duty and the platform setting are not the same thing.
- Remove synthetic imagery depicting games, interfaces or player outcomes that do not exist in the live product.
Google has not said whether gambling will follow elections into mandatory disclosure, and regulators may decide first. The ASA, Spain’s Directorate General for the Regulation of Gambling (DGOJ) and the Malta Gaming Authority (MGA) all police how gambling products are depicted, and none has set out a position on synthetic creative. The first ruling on an AI-labelled casino ad will tell operators more than the Google announcement does.
Source: Google
