Levi's
AI-generated models and product imagery
In March 2023, Levi's presented an AI-generated model pilot (Lalaland.ai) as a diversity lever, then had to publicly reframe within six days in the face of criticism about artificial diversity.
Objective
Test AI-generated models to show products on a wider variety of body types, ages, and skin tones, and let customers see themselves more represented.
The deployment
On March 22, 2023, Levi Strauss announced a pilot partnership with Lalaland.ai, an Amsterdam company that generates realistic AI models. The brand presented the move as a way to increase the diversity visible in its merchandising. The framing triggered immediate criticism: showing more diversity with synthetic people rather than with real models from underrepresented groups was perceived as diversity on the cheap, at the risk of pushing out professionals. On March 28, Levi's clarified that the pilot does not replace real shoots and is not a way to advance diversity.
Results Proof C
Established press (NBC News) and an academic case study (Journal of Contingencies and Crisis Management, Wiley) document the brand, the partner, and the statements by name. The brand's reframing statement supports the facts. No financial document, so C.
How it works
Inferred typical approachThe internal detail is not public. Here is a proven approach that leads to the same result, to adapt to your stack.
The stack in detail
- plateforme Lalaland.ai societe d'Amsterdam qui genere des mannequins realistes par IA, partenaire du pilote Levi's de mars 2023
- llm Moteur generatif Lalaland.ai modele proprietaire de generation de personnes synthetiques (morphologies, ages, carnations varies) ; architecture exacte non publiee
- infra E-commerce Levi's (fiches produit et imagerie) surface de diffusion prevue pour les visuels generes, aux cotes des shootings reels maintenus
Post-mortem
GraveyardWhat happened sourced
On March 22, 2023, Levi's announced a pilot with Lalaland.ai to generate AI models, presented as a way to increase visible diversity. Criticism grew within a few days: associations, press, and professionals accused the brand of simulating diversity instead of hiring real diverse models. On March 28, Levi's published a reframing: the pilot does not replace real shoots and is not a diversity instrument.
Reason for failure sourced
The mistake was in the framing, not the technology. Justifying an image generation tool with a diversity argument created a direct contradiction: replacing real underrepresented people with synthetic images runs counter to the stated goal. The brand had to separate the tool from its initial argument.
Cost sourced
Reputational cost: unfavorable press coverage for several days, the case cited as an example of a communication crisis. No direct financial cost published.
Warning signs inferred
Inferred: tying a tool that cuts visual production cost to an identity issue as charged as diversity was a risky communication bet. The confusion between operational efficiency and social commitment was predictable for anyone who tested the message with the affected audiences.
Lessons in hindsight inferred
Inferred: product image generation can be justified by display variety, scale, or cost, without being dressed up as social progress. The message must match what the tool actually does. On a sensitive subject, it is better to announce a practical and plain use than to claim a virtue that backfires.
Inferred: yes, AI-generated product imagery has become common since 2023 (virtual try-on, model variants, shoot variations). The failure is about the diversity framing of this announcement, not the category. The pattern holds if you are honest about what it brings: variety and scale, not real representation.
How your customers perceive this type of use
Sourced studiesUn ecart net separe les annonceurs des consommateurs : 77% des annonceurs voient l'IA positivement contre 38% des consommateurs (Yahoo/Publicis, 2024). Les mesures implicites confirment le rejet declare : en EEG, les pubs generees par IA produisent une activation memorielle plus faible que les pubs traditionnelles et sont decrites comme agacantes, ennuyeuses et confuses (NIQ, 2024). La disclosure a un effet ambivalent : elle augmente fortement la confiance quand elle est remarquee (Yahoo/Publicis), mais 27% des jeunes consommateurs disent faire moins confiance a une entreprise dont la pub est creee par IA (IAB, 2024).
Acceptance conditions
- Une disclosure visible : quand la mention IA est remarquee, la confiance globale envers l'entreprise augmente de 96% (Yahoo/Publicis 2024)
- Une qualite visuelle suffisante : les visuels IA de basse qualite augmentent l'effort cognitif et distraient du message (NIQ 2024)
Red lines
- Le contenu IA non declare puis identifie : 72% des consommateurs disent que l'IA rend l'authenticite difficile a etablir (Yahoo/Publicis 2024) et les marques utilisant des pubs IA sont plus souvent jugees inauthentiques ou non ethiques par les consommateurs que par les dirigeants (IAB 2024)
- Les mannequins et personnes generes par IA : 46% des consommateurs n'en veulent pas dans la publicite, l'inquietude premiere etant les standards de beaute irrealistes (Attest 2025)
Sources: Yahoo / Publicis Media (terrain Ebco) 2024 · IAB (avec Attest) 2024 · NIQ (NielsenIQ) 2024 · Attest 2025
How to replicate
Inference, not sourcedData prerequisites
- product catalog
- display rules per market
- brand guidelines
Org prerequisites
- legal and communication validation on the framing
- labeling of synthetic images
Possible stack
- model generation platform
- virtual try-on tool
- editorial review
First step: Define the tool's real argument (display variety, cost, scale) and ban any dressing up as social progress that would create a visible contradiction.
The same trap, elsewhere
CNET
SEO article generation at scale
Duolingo
Public communication of an AI-first strategy
generative AI response featured in advertising (false fact)
Sources
- S1 After backlash, Levi's says AI-generated models 'to increase diversity' won't replace real shoots Established press archive pending
- S2 Virtual supermodels made with AI spark fears in fashion workforce Established press archive pending
- S3 Levi's and Lalaland.ai collaboration crisis Secondary archive pending
An error, newer info, a source?
This page lives on its accuracy. If a figure has moved, if the deployment has changed, or if you have a higher-quality source, tell us. Every sourced correction is verified before publication.