AI-Generated Images in Ads: The Copyright Risks Every Advertiser Faces in 2025-2026

Using AI-generated images in advertising is fast, cheap, and visually appealing — but the legal risks are more serious than many marketers realize. In 2025 and 2026, courts, regulators, and lawmakers have drawn clearer lines around copyright ownership, infringement liability, and mandatory disclosure. Here is what advertisers need to know, backed by official sources.

The Core Problem: No Copyright for Purely AI-Generated Images

The single biggest issue for advertisers is that purely AI-generated images cannot be copyrighted in the United States. The U.S. Copyright Office has consistently affirmed that copyright protection requires human authorship . Its January 2025 report on AI and copyrightability concludes that "existing principles of copyright law are flexible enough to apply to this new technology," but that outputs created by generative AI are eligible only where a human author has determined sufficient expressive elements .

In March 2026, the U.S. Supreme Court declined to hear a challenge to this position, leaving the "human authorship requirement" firmly in place . This means that if your ad uses a purely AI-generated image — even one produced with detailed prompts — you may not have enforceable copyright against a competitor who copies it .

Even heavy prompting does not change this. The Copyright Office has explained that prompts (user instructions) do not provide sufficient human control over the expressive output to qualify as authorship . Copyright registration is only plausible when a human adds perceptible creative modifications — selecting, arranging, editing, or significantly altering AI-generated material .

Infringement Risk: The Image Itself May Violate Someone Else's Copyright

Even if you cannot copyright your AI-generated ad image, you can still be sued for using it. Generative AI tools raise infringement concerns in two ways: copyrighted materials may have been used in training, and the outputs themselves may resemble protected works . If an AI-generated image closely resembles a copyrighted character, artwork, or photograph, the advertiser using it may face infringement exposure — and the AI vendor typically is not liable .

In 2025, courts began confronting the merits of infringement claims against AI developers. There are now more than 80 active lawsuits targeting AI platforms and the agencies that use them . The Copyright Office's May 2025 report on AI training and fair use concluded that certain uses of copyrighted material for training "cannot be defended as fair use" . Advertisers should run thorough infringement checks on AI-generated imagery before publishing.

EU Transparency Rules: Mandatory Labeling Starting August 2026

For advertisers selling into the European Union, new transparency obligations take effect in August 2026. The EU AI Act requires providers of generative AI to ensure that AI-generated content is identifiable . Specifically, Article 50 requires that certain AI-generated content — including deepfakes and text published to inform the public on matters of public interest — must be clearly and visibly labeled .

Fines for non-compliance can reach up to 15 million euros or 3 percent of global annual turnover . This applies to any advertiser whose content reaches EU consumers, not just companies based in Europe.

EU Copyright: Human Creativity Still Required

The EU does not have specific Union-wide rules on the copyrightability of AI-generated works, but existing case law and Member State developments emphasize the need for human creativity . European Parliament materials state that "fully machine-generated outputs should remain unprotected, while AI-assisted works require harmonised protection criteria" . This mirrors the U.S. position: the AI Act is "silent" on copyright authorship, leaving the existing human-creativity framework in place .

Practical Takeaways for Advertisers

Advertisers can manage these risks with a few concrete steps:

• Do not rely solely on AI platform terms of service for copyright protection. Assess whether the image contains protectable human authorship under applicable rules .
• Add meaningful human creative input — significant editing, compositing, selection, arrangement, or original elements — if you want a stronger chance of copyright protection .
• Run infringement checks on AI-generated imagery before using it in ads. The risk is not only that you may lack copyright ownership, but also that the output may closely resemble protected third-party material .
• Label AI-generated ad content in the EU where required to comply with the AI Act's transparency requirements, which take effect in August 2026 .
• Stay current on evolving case law. More than 80 lawsuits are working their way through U.S. courts, and rulings continue to shape the liability landscape .