Major multinationals identify six key IP risks of GenAI in marketing

Please note that this report is WFA member only. If interested in WFA membership, get in touch with our team at membership@wfanet.org.

We have identified the key IP risks created by using GenAI in marketing for the first time in a new, practical report that also identifies how advertisers can tackle them.

Managing IP Risk when using Generative AI in marketing says the risks often come from inappropriate use of AI tools that can be managed with the right processes in place. The ambition is to allow WFA members to identify the right processes for their brands that will allow them to take advantage of the benefits of GenAI with reduced IP risks.

The six risks were identified via a process of qualitative research, involving the 10-member steering board for the WFA’s AI work as well as detailed responses from IP experts working within WFA member companies. The AI steering board includes representatives from AXA, Diageo, Essity, IKEA, Infosys, Kraft Heinz, Lego, L’Oréal, Mars and Teva Pharmaceuticals.

The six top IP risks for GenAI marketing are:

1. Loss of control over your brands’ IP and data input into GenAI tools by your staff, agency staff and other suppliers. Inputting data into an AI tool could mean it can be accessible to a wider audience.
2. Breach of third-party IP rights or confidentiality by inputting third-party IP or data into a GenAI tool without necessary permissions. Inputting insights into GenAI tools, for example, can expose brands to being in breach of their agreements with third-party providers.
3. Generating brand outputs that inadvertently infringe third-party IP rights. Images created by AI tools could, for example, closely resemble existing copyrighted or trademarked designs, leaving the brand open to legal action even if the resemblance was unintentional.
4. Lack of ownership or exclusive rights over IP in- and outputs created by GenAI tools. Rights to ownership of these brand assets might not be exclusive allowing other companies to use them in their own communications.
5. Third-party AI-generated fraudulent ads impersonating your brand in paid media. Brands can be put at reputational harm where GenAI is used to create me-too advertising that mimics logos and style. Such efforts often promote fake sales and direct consumers to counterfeit websites, damaging trust in legitimate brands.
6. Third-party organic content impersonating your brand. We’ve seen examples of AI-generated videos that “feature” celebrities and make claims about brands. When these spread on social media there can be a public backlash damaging the brand, alienating part of its customer base and forcing the brand to conduct damage control and clarify false claims.

The report, which also includes practical examples of how each of the above risks may surface, follows a quantitative study of WFA members, published in September 2024, which identified that 66% of brands state that legal challenges continue to be the main barrier to GenAI adoption, with 77% claiming they are most concerned about intellectual property (IP) and copyright risks.

Members were particularly concerned about using AI-generated content in marketing assets, with just 40% of those adopting GenAI today are using AI-generated content in user-facing marketing communications.

“Addressing GenAI’s legal risks is a top priority for brands. This report comes at a crucial time, providing brands with a clearer understanding of the IP challenges they face when using GenAI in marketing and actionable insights on how they could tackle them”, said Stephan Loerke, CEO of WFA.

The report also outlines potential actions that could be employed by marketers to reduce the risks above. Typically, these fall into one of three broad categories:

• Policies and processes: mitigations could focus on the potential policies and training that could be implemented to help ensure employees understand the IP risks of GenAI in marketing and how to use the technology responsibly.
• Contract and supplier management: action could focus on how contracts and relationships with AI providers, agencies and other partners could potentially be calibrated with a view to mitigating IP risks.
• Oversight and monitoring: these measures focus on the types of tracking and oversight tools that could be adopted to monitor the use of AI tools and identify any potential IP or copyright violations.

“Our goal is to empower marketers to unlock the technology’s potential as part of our steadfast commitment to providing its members with the knowledge and voluntary best practice they need to protect their brands and drive more effective and efficient marketing”, said Loerke.

WFA members can download the report here and an executive summary of the report here.