Generative AI is moving rapidly from creative experiment to everyday marketing tool. It can help brands produce content faster, personalise campaigns at scale and reduce production costs. But it is also creating a new category of commercial risk, particularly where advertising uses recognisable faces, voices or styles without proper permission.
For a business audience, that is the real issue. This is not just a question of legal theory. It is about brand trust, consumer confidence, reputational damage and the cost of getting a campaign wrong. Used well, AI can make advertising more effective. Used badly, it can trigger complaints, takedowns and unwanted headlines.
Whilst not related to the use of AI, in 2024 the Advertising Standards Authority banned paid-for ads by Simmer after edited footage made it look as though Sara Davies and Deborah Meaden from Dragons’ Den were endorsing its ready meals. The regulator found the ads misleading because consumers were likely to think the endorsement was genuine. It was a clear reminder that short-term attention can quickly become a compliance and reputation problem.
There are also examples from outside advertising that still matter to marketers. The viral AI song Heart On My Sleeve, which mimicked the voices of Drake and The Weeknd, was pulled from major platforms after attracting huge attention. More recently, OpenAI paused one of its chatbot voices after complaints that it sounded too similar to Scarlett Johansson. Both stories underline the same commercial point: voice and likeness now carry real economic value, and audiences are increasingly alert to how they are used.
Why this matters commercially
The commercial reality is straightforward. If an advert suggests a celebrity, creator or public figure supports a product when they do not, the fallout can be expensive. A business may face regulatory scrutiny, platform take-downs, wasted media spend, strained talent relationships and damage to customer trust.
Even where a claim does not end up in court, the disruption alone can outweigh any savings gained from using AI.
Brands, agencies and public figures are adapting quickly. Some celebrities are taking a more structured approach to protecting their personal brand through trade marks and tighter contractual controls. At the same time, marketing teams are building more approval steps into campaign workflows so they can check whether AI-generated content could imply a false endorsement, copy a recognisable style too closely or create a misleading impression for consumers.
It has long been recognised that a face, a voice, a catchphrase or even a familiar gesture can be a commercial asset. In a world of synthetic content, those assets need the same strategic protection as any other part of a brand.
Examples marketers should pay attention to
There are positive use cases as well. Fashion retailer Zalando used deepfake technology featuring Cara Delevingne (with her permission) to generate nearly 300,000 adverts specifics to various loca- tions, showing how AI can help scale campaigns efficiently when the necessary rights and permissions are in place.
Coca-Cola’s AI-generated Christmas advertising also showed the other side of the picture. The campaign generated major attention, but also criticism from viewers and creative professionals who felt the work lacked authenticity. Even where the legal position is manageable, there is still a commercial judgement to make about audience reaction and brand fit.
Public figures are also becoming more deliberate about protecting the elements that make them marketable. Taylor Swift has sought trade mark protection for aspects of her public persona, while sports figures such as Cole Palmer and Luke Littler have moved to protect distinctive branding linked to their image and identity. That matters because the stronger the commercial framework around a persona, the easier it becomes to challenge misuse.
The risk areas are familiar, even if the technology is new. Problems tend to arise where AI is used in ways that make consumers believe something untrue, such as:
- fake advertisements
- unauthorised endorsements
- cloned celebrity voices
- manipulated social media campaigns
- online scams
For advertisers, the consequences can include pulled campaigns, lost investment, negative press, complaints to regulators and awkward conversations with talent, rights holders or commercial partners. For individuals, it can mean reputational damage and loss of control over assets that have real commercial value.
What brands should be thinking about now
Most businesses are looking to take advantage of AI. For marketing teams, the practical question is not whether to use AI, but how to use it responsibly. In most cases that means checking rights and permissions early, being clear internally about approval processes, and pressure testing whether an advert could mislead consumers about who is involved.
Individuals or businesses looking to protect their brand assets or reputation should look at:
- auditing trade mark portfolios – consider unusual brand identifiers (like colours, gestures, sounds or slogans)
- updating endorsement agreements to address AI content
- monitoring for deepfakes and impersonations
- implementing rapid takedown procedures
It also means thinking about contracts and governance. Agreements with talent, influencers and agencies increasingly need to deal expressly with AI-generated content, synthetic voice use, editing rights and approval controls. That may sound operational, but it is now central to brand protection.
The bigger picture
AI is not going away, and neither is its appeal to advertisers and businesses in general. The businesses that will benefit most are likely to be the ones that combine experimentation with strong judgement and governance.
The legal framework will continue to evolve as well, the UK Government’s recent Report on Copyright and Artificial Intelligence specifically called out the issues with digital replicas and the potential for a new right to combat misuse.
If your business is using AI in advertising, or reviewing how to protect brand assets, endorsements and reputation in this fast-moving area, Howes Percival’s IP team can help.
By Stephen Ruse, Director in Commercial, Technology & IP at Howes Percival
For more information and advice, contact Stephen Ruse at stephen.ruse@howespercival.com or visit their website.
