AI Gains Ground in Marketing, Authenticity and Legal Risks Keep It Out of Final Creative

Generative AI in Marketing: Warm Adoption, Cold Feet on Human Likeness

Generative AI has made creative work faster. Drafts, scripts, code, and production tasks now move in hours, not weeks. But many brands still stop short of using AI for final, consumer-facing creative featuring people.

The gap comes down to authenticity, legal risk, and the "AI sheen" that turns audiences off. As one agency leader put it, "That missing layer of authenticity can create dissonance." For brands that trade on trust, that cost isn't worth the speed.

Where AI fits today

Most teams use AI for pitch materials, asset variations, and workflow automation. Think code optimization, asset resizing, transcription, and research. Liquid Death leans on AI for back-end efficiency, but keeps final creative analog unless the idea explicitly requires AI.

Columbia Sportswear follows a similar playbook. The brand sees value in scale and automation, but remains skeptical of AI-led consumer creative. "We know from social listening that Gen Zs reject AI content more than any other demo," said Matt Sutton, svp and head of marketing at Columbia Sportswear. The glut of "AI slop" hasn't helped.

The authenticity line

Expect backlash if the work feels fake or tries to pass AI off as real. Recent controversies around human-like imagery in fashion campaigns and even holiday ads show the risk is real. The trust tax is steep, and audiences are getting better at spotting AI gloss.

As Kate Wolff, founder and CEO of Lupine Creative, said, AI-generated human likeness in final spots isn't on the table. Until the uncanny valley is solved, brands will sit out or label AI clearly.

Legal heat is rising

The legal front is messy and active. Major media companies have sued AI vendors over alleged copyright infringement. Several publishers, including The New York Times, have filed suits against OpenAI. For context, see coverage from Reuters and the U.S. Copyright Office's guidance on AI-authored works here.

Agencies are shifting risk via indemnification to clients, production, or post. It's a hot-potato moment until case law settles.

Clear use cases that work

Brands can still win with AI-if they're transparent and original. Over the summer, Popeyes shipped an AI-generated diss track and labeled it upfront. The team cleared lyrics and likenesses and treated it like any high-stakes production. That's the bar.

Playbook: how to use AI without burning trust

• Define where AI helps: research, versioning, transcreation, QA, compliance checks, tags, and automation. Keep human likeness and final hero assets human-led unless you label them.
• Establish a "no AI sheen" standard: visual QA for hands, eyes, reflections, text artifacts, and implausible physics. Create a preflight checklist.
• Label AI clearly if it's part of the idea. Treat disclosure as a creative tool, not a legal shield.
• Get releases and rights in order: likeness, voice, music, datasets, and model terms. Build vendor questionnaires on training data and indemnity.
• Route legal early: approval paths, storage policies, and version logs. Document prompts and source material.
• Test with the audience: small cells, sentiment tracking, and social listening. Gen Z tolerance is lower-measure before you scale.
• Protect the brand: prepare responses for "Is this AI?" moments, including creator credits and process transparency.
• Train your team: prompt craft, review skills, and policy literacy. AI speeds output; your people protect taste and truth.

Bottom line

AI accelerates the process. Human craft earns the sale. Use AI to build smarter workflows and stronger pre-production, then ship work that feels real, reads true, and respects the audience.

If you're building an AI-ready marketing team, explore practical training and certifications for marketers at Complete AI Training.