Luxury Shoppers Are Using AI Faster Than the Brands They Buy From

Luxury shoppers in China and the United States are now using artificial intelligence to research, compare, and decide on high-end purchases at rates the brands they buy from have not matched. The June 30 study from Bain and Comité Colbert’s luxury AI study finds 64 percent of Chinese buyers and 54 percent of US buyers reported using AI tools during their most recent luxury purchase, against only 27 percent in France. The asymmetry between client behavior and brand-side rollouts is the larger story here. 82 percent of very heavy spenders used AI for their last luxury purchase, against 28 percent of light spenders, the same survey found.

Above that geography sits a commercial question. Bain’s separate 2026 luxury market forecast puts the lost active customer base at roughly 70 million since the post-pandemic peak of 2022, even as the global pool of affluent buyers kept expanding. The new study with Comité Colbert measures how much of that gap relates to who shows up in an AI search, and who does not. 97 percent of luxury AI users said they plan to use the technology again on an upcoming purchase, which is why the visibility question carries one-quarter weight rather than five-year weight.

Luxury Shoppers Have Outrun the Houses

Every Maison on Comité Colbert’s roster now lists artificial intelligence somewhere on its strategic agenda, the Bain study reports. 22 percent rank AI in their top three priorities for the next three years, against just 5 percent in 2024. Adoption has broadened across functions, mostly through pilots and tests rather than scaled deployments. Translating AI into measurable business impact remains the exception rather than the rule.

Bain frames the gap plainly: customers are moving faster than the industry. More than half of luxury consumers used AI during their most recent purchase in key markets including China and the United States. Strategy roadmaps inside luxury run on cycles longer than the AI discovery layer the consumer now occupies.

On the executive side, 39 percent of luxury groups and Maisons now report a defined AI vision, strategy, and sequenced roadmap, while another 48 percent have a vision paired with pilots. Multi-brand groups lead: all have a clear roadmap, befitting their role as central standard-setters, and Maisons within those groups benefit from the corporate momentum. Most independent Maisons outside the largest remain a step behind, constrained by limited internal resources and a reliance on external partners. Roughly 60 percent of luxury companies have yet to realize significant impact from any AI deployment, the study finds. Even in the more advanced areas, like knowledge management and IT, only 13 to 15 percent of respondents say those deployments have generated significant business impact.

AI Adoption Splits by Country and by Spending Tier

The headline asymmetry becomes a tier signal once you break it down by country. In April 2026, Bain surveyed 534 French, 559 US, and 512 Chinese luxury consumers about AI use during their most recent luxury purchase. The country gap is wide: 64 percent of Chinese buyers, 54 percent of US buyers, and 27 percent of French buyers said they used AI to help. The behaviour also crossed the store wall, with 47 percent of in-store luxury buyers reporting AI use during their shopping journey.

Above country, spend level is the sharper divider. 82 percent of very heavy spenders used AI for their most recent luxury purchase, against 51 percent of moderate spenders and 28 percent of light spenders. The study reads that pattern as a value signal rather than a novelty one: customers with more at stake are quicker to lean on external research. The behaviour runs across product categories too, with AI use ranging from 55 to 58 percent regardless of what was bought. Near-universal satisfaction closes the loop: 97 percent of luxury AI users plan to use the technology again for an upcoming purchase.

Source: Bain AI consumer survey, April 2026.

Generative Search Is the New Luxury Shelf

Discovery now starts inside large language models. About 75 percent of luxury prompts carry a discovery or comparison intent, precisely the stages where a maison has the most to gain from a deliberate strategy. The discipline has a name in the report: generative engine optimization, shortened to GEO. Most organizations have only begun to address this shift. 70 percent of luxury-related queries do not mention any brand name, so only 30 percent of consumers enter a generative search with a specific brand in mind.

Joelle de Montgolfier, executive vice president of global retail and luxury at Bain & Co., puts the consequence plainly in the report. The engines choose which brands consumers discover, since most prompts arrive without one in mind. The maison absent from that shortlist never enters the comparison set. Discovery without a brand name in mind starts at the model layer, not at the search box.

The engines draw on an information ecosystem that looks nothing like classic search. 90 percent of the citations those large language models surface come from off-site sources, the study finds, with editorial coverage, customer reviews, blogs, and resale platforms doing most of the loading. Official brand domains account for only 10 percent of cited sources for watch queries and 45 percent for jewelry. 60 percent of Maisons and groups actively work on the content and structure of their own sites. Only 26 percent address off-site content, which is where the engines mostly read.

Small Specialist Brands Now Beat the Houses in AI

Among the 30 most visible luxury brands on LLMs, only 23 percent are large corporations with revenues above €5 billion, Bain’s analysis with the GEO firm Meikai finds. 70 percent of those large Maisons capture less visibility than their revenue share, while most mid-sized Maisons over-index. Every small Maison in the group outperforms its market weight in visibility by three to eight times. Reviews and third-party editorial do most of the loading in unbranded prompts.

In watches and skin care, specialist brands are referenced more often than diversified luxury groups, even at a fraction of their sales. 60 percent of groups and Maisons actively work on the content and structure of their own sites. Only 26 percent address off-site content, the same study finds. The biggest owned-site investments do not always translate into top citations on LLMs.

Tracking the new shelf is also lagging. 48 percent of groups and Maisons regularly track their AI search and GEO performance. 52 percent rate their visibility as moderate and 29 percent as weak. Only 10 percent consider themselves strong on generative visibility.

Luxury’s Approval Cycle Is the Real Bottleneck

The brand-side slack is not a strategy problem. Nathalie Remy, senior partner for retail, fashion and luxury at Bain & Co., describes a culture of perfection inside luxury that makes slow AI rollouts structural rather than a matter of will. The bottleneck sits in the approval stack, not in the strategy room.

By the time we’ve developed a tool, secured legal approval, image approval, and every internal signoff, the technology has already moved on.

Remy tied the remark to a wider luxury executive consensus in the report. The same hesitation shows up inside luxury hiring, where heritage houses are slowing AI adoption over brand voice concerns. Karen Harvey, CEO of the luxury advisory firm Karen Harvey Consulting, told Vogue Business earlier in 2026 that luxury houses will be much slower with AI adoption than young people would expect, with creatives at heritage maisons worried their work will be watered down. That caution does not show up at digital-first retailers chasing throughput.

The trade-off is the visibility gap. 97 percent of luxury AI users said they plan to use the technology again on an upcoming purchase, a near-universal satisfaction score. Brand websites, once the canonical reference, are now one input among many. The longer a maison waits, the more the engine cements which brands it surfaces by default. Remy argues the response should be a smaller number of meaningful AI initiatives, measured and iterated, rather than endless pilots.

What Luxury Houses Need to Build

Bain closes the study with a six-step GEO checklist the industry can act on this quarter. The first three priorities sit at the strategy layer: pick a territory, align content to the way consumers phrase queries, and weight off-site work over owned-site work. The next three sit at the operations layer: curate reviews, keep content fresh, and make sites readable to the engines. Each step is a project, not a slogan.

Bain’s warning on cadence is pointed. Remy, in the report, argues Google had no expiration date while large language models do. The checklist works only if the maison sustains it, since continuous monitoring and adaptation close the loop.

1. Claim a territory. Determine where you want to be a visible reference, then ensure the narrative is distinct, consistent, and uniformly deployed across every online touchpoint crawled by LLMs.
2. Align content to shopping intent. Use the language consumers actually phrase their queries in, and reflect both the stage of the customer journey and the buyer’s intent.
3. Prioritize off-site content. On LLMs, third-party citations carry more weight than a brand’s own copy, so cultivate online communities, social platforms, marketplaces, and specialized media.
4. Curate consumer reviews, since sites rich with reviews are cited more often.
5. Maintain fresh content. LLMs favor recently updated sources, both on-site material and third-party content.
6. Optimize sites for LLM readability with indexable pages, structured metadata, clear semantics, and FAQs, and add continuous tracking and measurement that adapts content as GEO evolves.

Frequently Asked Questions

How many luxury buyers used AI on their last purchase?

Sixty-four percent of Chinese buyers and 54 percent of US buyers said they used AI during their most recent luxury purchase, against 27 percent in France, per a Bain and Comité Colbert survey of 1,605 consumers run in April 2026. Among very heavy spenders, the figure reaches 82 percent.

What is GEO, and why does it matter for luxury?

Generative engine optimization, shortened to GEO, is the discipline of earning visibility in AI-generated search answers. Bain finds 70 percent of luxury-related AI prompts do not mention any brand, so engines largely decide which maisons a buyer first encounters. 90 percent of the citations those engines draw on come from third-party sites, not from brand-owned pages.

Why are smaller luxury brands showing up more on AI than giant houses?

Bain’s analysis shows that among the 30 most visible luxury brands on LLMs, only 23 percent are large corporations with revenues above €5 billion. 70 percent of large Maisons capture less visibility than their revenue share, while every small Maison in the same group outperforms its market weight in visibility by three to eight times. Reviews and editorial coverage, the inputs where small specialists often over-index, drive most LLM citations.

What is the biggest reason luxury brands have not caught up with AI?

The study points to internal approval cycles and a culture of perfection. Legal, image, and sign-off processes run longer than the technology cycle. By the time a maison clears a tool, consumer behavior and AI capabilities have already moved on, Remy argues in the report.