Dior’s AI Dispute Shows Why Advertising’s Binary Labels Fail

Three outlets had asked Dior about generative-AI claims by Friday, and the luxury house had not answered.

Dior’s new animated holiday short, “Night in Color,” became the focus last week after viewers spotted a cartoon character with one extra joint, warped lettering and a lazy eye. Online commenters concluded that generative AI had made the film, a conclusion brands often promote as a marketing achievement.

Jonathan Anderson, Dior’s creative director, had taken the opposite position. He told Women’s Wear Daily that the film was “all hand-drawn.” The claim placed costly human craft at the center of a luxury campaign and made production method part of the advertisement.

Sylvain Chomet directed the talking-animal fairy tale, which places Christian Dior and Marlene Dietrich in an animated story. Dior also released a making-of film to show the production craft. The work itself was effective, which made the provenance dispute more consequential than a technical footnote.

Ben Wild, who runs a film studio whose clients include Netflix and Disney, said two people claiming to have worked on the film told him its motion was largely AI-generated, while the furniture and bees were entirely AI-generated. Dior has not confirmed Wild’s account. Nor has it denied it.

Director Chomet denied the use of generative AI, but on Friday acknowledged its use to smooth the animation. The distinction leaves room for machine-assisted finishing without confirming Wild’s stronger account of the motion, furniture or bees.

How far does AI now reach into media production?

Almost nothing people watch or read is now free of machine assistance. The practical shift is not from human work to machine work; it is toward a mixed production chain that rarely fits either label.

• A phone blends several frames before a photograph is finished.
• Photoshop’s generative tools can invent the missing half of a sky.
• AutoTune has machine-assisted pop vocals for more than 25 years.
• Video captions are automatic, while machine learning cleans dialogue in edit suites.
• Claude checked every sentence of the commentary behind this debate for grammar and factual errors.

Around 2024, the clean divide between real and artificial production gave way to a continuum. Work now runs from slight noise reduction to a creature imagined and drawn entirely by machine. No honest binary label captures that range.

The useful questions are where a project sits on that gradient and whether the brand says so. A generated bee may not matter. A false statement that no AI was used does, especially when human creativity and hand-made identity form the brand’s provenance.

Why did the hand-drawn claim become the campaign?

In a feed crowded with machine-made video, visible and fallible human effort has become a luxury cue. Prada, Porsche, Levi’s and Apple also emphasized manual production for the same reason. Dior wanted that same hand-drawn distinction, but the cue works only when the process carries enough time and cost to look credible.

Wild’s version is that Dior rushed into the film after hearing that Prada had an animation legend available. His account points to a simple constraint: work of this kind cannot be accelerated without AI.

Had Dior said early that artists drew the keyframes and a machine filled the gaps, a few purists might have complained. The wider audience could still have admired the bees and appreciated that Dior was a modern brand using the tools available to it. Instead, Dior made purity the claim, and purity became the campaign’s weak point.

Anderson’s central idea was refreshing, and the film worked as advertising. The dispute was largely avoidable. It attached a production-method argument to a campaign whose appeal came from the work itself.

What does this change for luxury marketers?

For CMOs at luxury brands, the practical change is straightforward: disclose the real production mix and do not market a binary label. Both “AI generated” and “entirely hand-made” are becoming impossible positions to defend.

A mass brand can often absorb a credibility gap because audiences have no expectation that a Doritos ad was drawn by monks. Luxury works differently. It sells provenance first and product second, so provenance is the product.

When a story about human hands turns out to be about graphics cards, the damage exceeds one advertisement’s reception. It invites customers to reconsider how they think about the brand, including the signals they pay a premium for.

The broader lesson is that advertising’s AI-or-not argument now offers little value. Treating the method as a boast or a purity claim gives the process more importance than it can bear. The work is the work; the production claim should support it, not become the story.