AI in Luxury Isn't All-or-Nothing
What's genuinely worth a mid-sized maison's attention, and what's expensive theatre
The AI conversation in luxury is dominated by two extremes. On one side, conglomerates running agentic AI programmes on budgets larger than most maisons' total revenue. On the other, a Vogue advertisement that put its own readers against it. Neither extreme is useful guidance for a mid-sized maison deciding what to actually fund this year. Most of the real answer sits quietly in between, and it isn't complicated once you separate what's genuinely worth doing from what's theatre.
Start with why this isn't optional anymore. Bain and Altagamma's latest luxury market analysis finds that roughly half of luxury consumers already use AI somewhere in their purchase journey: a quarter for product discovery, two-thirds for comparing options before they buy.¹ The report's own framing is blunt: brands that aren't building “AI-native relevance” risk being left out of how their own customers now shop. The question was never whether to engage with this. It's which parts.
Worth funding
Being findable where the comparison happens. If two-thirds of your buyers are using AI to compare options before they choose, the maison needs to be accurately and consistently represented wherever that comparison happens: clean, structured product information, consistent facts across every channel and listing. This is unglamorous and inexpensive, and it's the direct AI-era descendant of the single-source-of-truth problem I've written about before. Get the underlying data right and the visibility follows; skip it and no amount of marketing spend fixes being invisible or misquoted in an AI-mediated search.
Provenance and authentication. Global trade in counterfeit and pirated goods runs to roughly $467 billion a year by the OECD's own count, and luxury goods account for around 60% of it, the category hit hardest.² The AURA Blockchain Consortium, founded by LVMH, Prada Group, Cartier and OTB, now offers a no-code SaaS version of its authentication and traceability platform built explicitly “with smaller brands in mind,” with lower onboarding costs and no in-house blockchain team required.³ This is the rare case where the technology matured faster than its reputation: it's no longer a conglomerate-only tool.
An advisor who actually remembers the client. Ralph Lauren's “Ask Ralph” shows the shape of the idea: a conversational assistant that recommends real, in-stock pieces based on what the client tells it in the moment, with deeper personalisation to the individual customer already on Ralph Lauren's own roadmap.⁴ A mid-sized maison doesn't need Ralph Lauren's version, or Ralph Lauren's budget. It needs the underlying capability Ask Ralph is still building toward, at boutique scale: a sales advisor who can see a client's full history across boutique, DTC and wholesale before the conversation starts, not a chatbot that replaces her. That's a data problem before it's an AI one; solve the first and the second becomes a modest addition, not a project.
Worth skipping, for now
Generative imagery instead of your own creative work. When Vogue ran a Guess advertisement with an AI-generated model, the backlash wasn't really about the technology. It was about what the image implied the brand no longer valued: the photographer, the stylist, the craft of making the picture.⁵ For a mass brand, that's a bad news cycle. For a maison whose entire commercial premise is craftsmanship and authenticity, the same shortcut spends the one asset you can't easily rebuild. The saving is real. The risk is disproportionate to the size of the brand taking it.
Building your own AI infrastructure. Proprietary agents, in-house models, a data science team to run them: that's a conglomerate's balance sheet, not a founder-controlled maison's. Almost every capability above this line is now available as a platform or a consortium membership rather than something you'd build. Buy the capability; don't staff the R&D.
The pattern underneath both lists is the same. The good investments make the maison more findable, more verifiably genuine, and better-informed about the client it already has. The risky ones try to manufacture, synthetically, the one thing luxury has never been able to outsource: the relationship, and the craft behind it.
If you're weighing which of these actually applies to your house this year, I'm always glad to compare notes.
Elisabeth
The Beauty of Data. Designed for Growth.
References
1. Bain & Company / Altagamma, global luxury market report and press release, 2026 (AI usage in the luxury purchase journey; “AI-native relevance”). https://www.bain.com/about/media-center/press-releases/2026/global-luxury-stabilizes-amid-compounding-disruptions-as-brands-race-to-amplify-meaning-and-rebuild-relevance/
2. OECD/EUIPO, “Mapping Global Trade in Fakes 2025” (global trade in counterfeit and pirated goods, and the luxury-goods share of it). https://www.oecd.org/en/publications/mapping-global-trade-in-fakes-2025_94d3b29f-en.html
3. Prada Group, “Aura Blockchain Consortium Launches Aura SaaS for Luxury Brands.” https://www.pradagroup.com/en/news-media/press-releases-documents/2022/22-01-13-aura-saas-luxury-brands.html
4. Ralph Lauren Corporation, “Ralph Lauren Introduces Ask Ralph, a New Conversational AI Shopping Experience.” https://corporate.ralphlauren.com/pr_250909_AskRalph.html
5. FashionNetwork, “Vogue US faces backlash over Guess ad featuring AI-generated model.” https://us.fashionnetwork.com/news/Vogue-us-faces-backlash-over-guess-ad-featuring-ai-generated-model,1754907.html
