Everyone is focused on the wrong part of agentic commerce. The Walmart + OpenAI
By Destaney Wishon · March 21, 2026 · Curated by George's Blog
Everyone is focused on the wrong part of agentic commerce.
The Walmart + OpenAI checkout failure isn't proof that AI agents won't disrupt ecommerce.
Checkout is a mechanical problem. It will get solved. That's the easy part.
The real disruption that brands and retailers should be worried about is happening at the discovery level.
If customers start relying on AI tools to find and recommend products for them, generic search traffic declines, and so does digital shelf exposure for every SKU in that category.
Retail Media Networks are built on the assumption that brands need to pay for visibility inside a retailer's walled garden.
But AI agents carry a depth of context about a shopper that no RMN can compete with on its own.
When a shopper's agent already knows their preferences, past purchases, budget, and lifestyle, and for me....even my blood work, why would I start a search on Amazon or Walmart at all?
Brands that have relied on retailers to drive demand will see a major decrease in share. Especially those whose primary driver of discovery is generic category terms, which are already shrinking as branded search grows.
AEO isn't the same thing as a shopping agent, but it is the catalyst that will fundamentally change how shoppers interact with platforms, with or without seamless checkout.
Fixing checkout is an engineering problem. Becoming invisible before the shopper even starts looking is the actual disruption.