Artificial intelligence is transforming how ads are created, styled and delivered - but when it comes to compliance, there’s no shortcut, no cheat code, and no “auto‑fix” button. Whether you’re experimenting with generative imagery, deploying automated ad platforms, or tempted to plug a celebrity likeness into your next campaign, the CAP Code still applies.
Here are four key things advertisers need to know before they compile, render or deploy AI‑generated content.
-
The CAP Code is media‑neutral - no exceptions, no escape keys
AI might change how ads are made, but it doesn’t change how they’re regulated. The Code is media‑neutral, meaning the rules apply regardless of how the content was created. If an ad falls within the ASA’s remit, it must comply with all relevant rules.
The ASA’s ruling on Polyverse Inc demonstrates that harmful, offensive or socially irresponsible AI‑generated images can breach the Code just as any other image would.
-
Deepfakes don’t get a free pass on misleadingness or endorsements
If an AI‑generated celebrity “endorsement” looks real enough to mislead consumers, it’s likely breaching the Code.
And while personality rights and intellectual property aren’t issues the ASA regulates, advertisers should still consider the legal risks of using a public figure’s likeness without permission and seek legal advice. Just because a tool can generate a convincing deepfake, doesn’t mean you can safely use it.
-
AI bias isn’t just a tech problem - it’s a compliance problem
In the Cosmos Oyun Yazilim ruling, in-game ads for an AI simulator app that objectified women breached the Code.
If an AI tool produces something problematic, the advertiser remains responsible for any output used in their ads. They should therefore sense‑check AI outputs to ensure they are:
- Socially responsible
- Free from harmful, discriminatory stereotypes and misleadingness
- Appropriate for the audience and placement
-
Automated campaigns don’t remove responsibility
The ASA’s Stripe & Stare Ltd ruling is a key reminder that even when ads are generated or distributed through automated AI platforms, responsibility does not shift.
The ASA found that an image of a model with an unbuttoned top presented her as a stereotypical sexual object - something not required to advertise the product. The ad was ruled irresponsible, particularly given its placement on a gaming website, which was poorly targeted.
Crucially, the ASA stated that even though the ad was produced using Google’s automated tools, the advertiser was still primarily responsible for ensuring compliance with the CAP Code.
AI can be a powerful creative tool, but advertisers must always sense‑check outputs and remember that they are responsible for the final result.
In short…
Before you hit ‘export’, make sure your AI‑assisted campaign isn’t about to generate more complaints than conversions.
Anyone looking for advice on their AI-assisted campaigns are invited to contact our very real and human Copy Advice team.
More on
-
Keep up to date
Sign up to our rulings, newsletters and emargoed access for Press. Subscribe now.
