For years, brands have been trying to solve the same distribution problem: how to benefit from powerful platforms without becoming dependent on them.
Amazon gave brands access to enormous purchase intent, but also placed them inside an environment where price, reviews, availability and sponsored visibility could matter more than the relationship between the customer and the brand. Google became critical to product discovery, while Instagram and TikTok created entirely new ways for brands to build demand. Each platform created growth opportunities, but also gave another company meaningful influence over how that growth was distributed.
This helped drive the rise of direct-to-consumer strategies, owned audiences, first-party data, loyalty programs and brand-owned ecommerce. Companies increasingly wanted both reach and control.
Now AI shopping could introduce another intermediary, and potentially a more influential one.
The difference is that AI assistants may not simply determine what consumers see. They could increasingly influence what consumers choose.
From Search Engines to Decision Engines
Traditional search gives consumers options. A shopper searching for running shoes, skincare or luggage receives pages of products, retailers, advertisements and reviews. Google helps organize the market, but the consumer still does much of the comparison.
AI shopping changes that interaction.
Instead of searching for "best carry-on luggage," a customer can explain exactly what they need: they travel internationally twice a month, carry a laptop, prefer something lightweight and have a $400 budget.
The AI can interpret those requirements, compare available products and reduce hundreds of possibilities to three or four recommendations.
As agents become more capable, another step could disappear. The customer may eventually ask the assistant to choose the retailer, find an available offer and complete the transaction.
That moves AI from being another discovery channel toward becoming a decision layer between brands and consumers.
Brands Have Already Lived Through One Platform Dependency Cycle
Amazon provides perhaps the clearest precedent.
For thousands of consumer brands, being available on Amazon provides enormous distribution. But Amazon also owns the shopping environment, controls the interface and sits between the merchant and the customer.
That trade-off helped make direct commerce strategically attractive.
Nike provides an interesting example of the broader thinking behind this shift. Over the past decade, the company invested heavily in its own apps, membership ecosystem and digital commerce while reassessing its relationships with wholesale partners. The underlying logic was bigger than ecommerce margins. Direct relationships gave Nike greater access to customer behaviour, product preferences and loyalty.
The broader lesson is not that brands should abandon platforms. Nike itself demonstrates why that conclusion would be too simplistic as its distribution strategy has continued evolving.
The lesson is that distribution reach and customer ownership are different things. AI shopping could force brands to confront that distinction again.
The Next Marketplace May Not Look Like a Marketplace
Amazon looks like a marketplace. Instagram looks like a social network. Google looks like a search engine.
The next major commerce intermediary could simply look like a conversation. A consumer might ask an AI assistant:
"Find me a pair of running shoes for marathon training under $200. I prefer something lightweight and don't want an aggressive carbon plate."
The assistant could compare products from Nike, Adidas, New Balance, ASICS, On and dozens of smaller brands before presenting a handful of recommendations.
From the consumer's perspective, this is convenient.
From the brand's perspective, something important has happened: the AI has determined the consideration set.
The consumer may never visit most of the brands that were evaluated. They may never see their homepage, campaign creative, product navigation or carefully designed ecommerce experience.
Those brands effectively competed for the customer before the customer ever arrived.
Retailers Are Already Preparing for This Shift
This is not purely theoretical. Large retailers including Walmart, Ulta Beauty and Wayfair have been adapting their digital properties and product information to improve how their products appear in AI-assisted shopping environments.
That tells us something important about where ecommerce infrastructure is heading.
Historically, retailers optimized product pages for two audiences: consumers and search engines.
Increasingly, there is a third audience: AI systems trying to understand products well enough to recommend them.
A Wayfair product, for example, cannot rely exclusively on an attractive product page. AI systems need to understand dimensions, materials, style, price, availability, reviews and the contexts in which that product might be appropriate.
The digital shelf is becoming machine-readable.
Shopify Shows the Other Side of the Equation
Shopify became strategically important partly because it gave brands an alternative to building their businesses entirely inside marketplaces.
A merchant could own its storefront, customer experience and customer data while still using external platforms for acquisition.
AI commerce creates a new version of that challenge. The storefront can remain owned by the brand while discovery increasingly happens elsewhere.
A consumer could discover a Shopify merchant through an AI assistant, receive a product recommendation there and only reach the merchant when it is time to transact.
In that world, the website does not disappear. Its role changes. It becomes less responsible for initiating every customer journey and more responsible for converting externally generated demand into a relationship worth keeping.
The Bigger Risk Is Losing the Customer Relationship
Payments company Adyen has highlighted exactly this tension. As AI-assisted shopping develops, merchants are thinking harder about how much demand comes through direct channels and how they preserve those relationships if large language models become more influential sources of commerce.
That concern goes beyond transaction fees. The customer relationship contains purchase history, preferences, loyalty, service interactions, memberships, personalization and future purchasing potential.
Imagine a customer saying: "Reorder my moisturizer."
For a company such as Estée Lauder, L'Oréal or a digitally native skincare brand, the important question is not simply whether its product is available to the agent.
The question is whether the consumer has enough preference for the brand that the AI is instructed to reorder that specific product, rather than reassessing the category and recommending something else.
That distinction could become enormously important.
AI Could Make Brand Preference More Valuable
There is an argument that AI will weaken brands because intelligent systems can compare products more objectively. But AI could also make strong brands more valuable.
Consider the difference between two prompts: "Find me the best running shoes for marathon training." and "Find me the best Nike shoes for marathon training."
In the first scenario, Nike is competing with Adidas, ASICS, New Balance, On, Hoka and potentially dozens of alternatives. In the second, Nike has already won the most important part of the decision.
The customer has constrained the agent. This is why brand building could become more important in an AI-mediated commerce environment, not less. The objective is not merely to become something an AI system recommends. It is to become something consumers explicitly ask their AI systems to choose.
Loyalty Could Become Infrastructure
This also changes how brands should think about loyalty. Historically, loyalty programs have often been treated as retention mechanisms: points, discounts, rewards and occasional offers designed to encourage another purchase.
AI commerce could make loyalty much more strategic. Nike Membership, Sephora Beauty Insider, Starbucks Rewards and similar ecosystems create reasons for customers to maintain identifiable relationships with specific brands or retailers.
Those relationships can include purchase history, personalized recommendations, exclusive access, services, rewards and accumulated benefits that cannot easily be replicated by an AI assistant switching merchants for every transaction.
In that environment, loyalty becomes more than a marketing program. It becomes protection against interchangeability.
AI Visibility Is Only Half the Strategy
Brands therefore face two simultaneous challenges. They need to become understandable and recommendable inside AI systems. Product information, structured data, third-party authority, reviews, availability, reputation and broader digital presence will increasingly influence whether AI systems can confidently include them in recommendations.
But brands also need to ensure that AI-generated discovery eventually strengthens rather than replaces their direct customer relationships.
This creates a different way of thinking about AI visibility. The goal should not simply be: Get recommended by AI. The better objective is: Get recommended by AI, convert that recommendation into a customer, and give that customer a reason to come back directly.
The Next Distribution Moat
The internet has repeatedly rewarded brands that understood new distribution platforms early. Google created search-native companies. Amazon created marketplace-native brands. Instagram and TikTok created social-native brands. Shopify helped make direct-to-consumer commerce dramatically easier.
AI assistants could create the next major distribution layer. But brands should remember what previous platform cycles taught them. Distribution can be borrowed. Customer relationships are much harder to replace.
The strongest brands will therefore not choose between AI distribution and direct commerce. They will learn how to use AI assistants for discovery while strengthening the parts of the customer relationship they can actually own.
That could become one of the defining principles of commerce in the AI era: Be discoverable everywhere without becoming dependent on anywhere.
