AI Shopping Agents Retail: Why Walmart's Gateway Changes Everything

On October 21, Walmart CEO John Furner stood on stage at the Masters of Scale Summit and announced something that will reshape how every consumer brand thinks about distribution: AI shopping agents retail access is no longer optional, it is infrastructure. Walmart is opening its stores, both Walmart.com and Sam's Club, to third-party AI agents that can browse, build carts and check out on a customer's behalf. Amazon, the company that effectively invented one-click commerce, is doing the opposite and locking agents out entirely.

This is not a minor product update. It is the starting gun on a new competitive era where the customer standing between a brand and a sale might not be a customer at all, but a piece of software negotiating on someone's behalf. Matt Britton, founder of Suzy and one of the country's leading AI keynote speakers, has been telling Fortune 500 audiences for the past year that the next disruption in consumer behavior would not come from a new app or platform. It would come from the disappearance of the human decision-maker inside the purchase funnel.

The numbers back him up. Bain projects the U.S. agentic commerce market will reach $300 billion to $500 billion by 2030, representing 15% to 25% of total ecommerce sales. Gartner separately forecasts that 20% of digital commerce transactions will run through AI platforms by the same year. Kearney research shows 73% of consumers are already familiar with AI shopping tools, and 60% expect to use an AI agent for shopping within the next 12 months.

Britton's core argument is simple and urgent: the brands that win the next decade of retail will be the ones that build trust and visibility into the AI agent layer right now, not after the holiday shopping season proves the point for them. Walmart just made that choice for every brand on its platform. Amazon just made the opposite choice. CMOs now have a live case study in real time, and this post breaks down exactly what it means for marketing, pricing and brand loyalty going forward.

What Walmart's Agent Gateway Actually Does

Walmart's announcement introduced three connected pieces of infrastructure. Agent Profiles let shoppers authorize AI agents, set spending limits and track purchases, Agent Gateway gives agents a more direct connection to Walmart's shopping services, and new purchasing rules set limits on agents' actions and access to customer information. Together, these tools function like a corporate card program, except the cardholder is a piece of software acting on a human's behalf.

The mechanics matter for anyone trying to understand agentic commerce as a business model rather than a buzzword. Walmart will let AI shopping agents buy on a customer's behalf through separate Agent Profiles, with purchase approval by default or a spending cap the customer sets, and agents never see card details and cannot change payment methods. Shoppers can choose a cap like $50, $100 or $200, or simply require approval on every single purchase.

The technical backbone is just as important as the consumer-facing controls. The Walmart Agent Gateway is an MCP server in beta for developers, running on the Model Context Protocol and letting AI agents search, build carts and check out without browser automation. That last detail is the one most executives miss. Instead of an agent clumsily clicking through a web page like a human would, it connects directly to Walmart's systems through a structured, machine-readable interface built specifically for this purpose.

Walmart also built in guardrails that protect both the shopper and the brand experience. Agents on Walmart's site see the same prices a human would, with offers and fees shown before purchase, and unintended agent purchases go through Walmart's normal returns process. That price parity commitment is a deliberate signal to regulators and skeptical consumers that the retailer is not trying to exploit the agent layer for margin extraction.

Why Amazon Chose the Opposite Path

While Walmart was building a welcome mat, Amazon was reinforcing the walls. Amazon, a month earlier, had blocked Meta's Muse agent from shopping its site. The company's stated rationale centers on control and liability. Amazon blocked third-party agents citing lack of authorization to access customer accounts, scrape data, or process transactions.

Amazon's resistance goes beyond one blocked integration. Amazon also sued Perplexity last year over its Comet browser's shopping agent, and that legal battle is still ongoing after a court refused Amazon's request to ban the tool in August. The strategic logic is not hard to decode. Amazon would rather shoppers use its own agent, Alexa for Shopping.

That bet carries real risk. If an outside agent reads a store's catalogue, picks the SKU, and delivers the shopper to checkout, the margin from merchandising, sponsored listings and recommendation surfaces evaporates, and for a store whose business model monetizes attention inside the browse experience, letting agents in without a value exchange is a self-inflicted wound. Amazon has built a trillion-dollar advertising business on top of human browsing behavior. Letting outside agents bypass that browsing layer threatens the exact revenue stream Amazon has spent two decades perfecting, which explains the lockdown far better than any stated security concern.

Matt Britton frames this split as a preview of a much larger industry fracture. Every major platform, from travel sites to marketplaces to direct-to-consumer brands, will soon have to pick a side: open the gateway and compete for algorithmic attention, or wall off the garden and bet that customers stay loyal to a single branded assistant. Neither choice is risk-free, but only one of them is reversible once agent-driven shopping habits take hold.

AI Consumer Behavior 2026: The Loyalty Problem Nobody Has Solved

The deepest disruption here is not technical, it is behavioral. For decades, brand marketing has operated on the assumption that a human being sees an ad, forms a preference and clicks "buy." That chain is breaking. When an agent is instructed to find the best reclining sofa or the cheapest 20-foot trailer, the brand's job shifts from winning a human's attention to winning an algorithm's selection logic.

This is the exact dynamic Matt Britton explores in his book Generation AI, which argues that the generation coming of age right now will default to AI-mediated decisions for everything from entertainment to groceries. Britton's research through his consumer intelligence platform, Suzy, has tracked a steady decline in deliberate brand search behavior among younger shoppers in favor of delegated, assistant-driven discovery. That shift does not pause for the holiday shopping calendar, it accelerates with it.

The pricing implications are just as unsettled. A recent Bloomberg Odd Lots episode examined whether agentic shopping actually benefits consumers or simply automates a new form of price discrimination. Research from the think tank Groundwork Collaborative shows that a number of consumer facing agents end up serving customers more expensive items after learning more about their financial situation. Owens, the guest on that episode, argues that the history of price gouging does not bode well for claims that things might get cheaper because of new technology.

That finding should worry every CMO currently treating agentic commerce as a simple efficiency play. If agents can be nudged toward higher-margin products based on a shopper's perceived ability to pay, then "conversion" stops meaning what it used to mean. Brands now need visibility into how they are represented, ranked and priced inside an agent's recommendation logic, not just how they rank on a search results page.

Agentic Commerce by the Numbers: What the Market Scale Tells CMOs

The scale of this shift is already measurable, and it is moving faster than most retail roadmaps account for. Consider the trajectory of market forecasts from the last twelve months:

• $300 billion to $500 billion: Bain's projected size of the U.S. agentic commerce market by 2030, representing 15% to 25% of total ecommerce sales.
• 20%: Gartner's estimate of digital commerce transactions that will run through AI platforms by 2030.
• 73%: Share of consumers already familiar with AI shopping tools, according to Kearney research.
• 60%: Share of consumers who expect to use an AI agent for shopping within the next 12 months, per the same Kearney study.
• $50, $100, $200: The spending-limit tiers Walmart is rolling out through Agent Profiles, giving shoppers a built-in budget ceiling for autonomous agent purchases.

These figures tell a consistent story. Agentic commerce is not a speculative future state, it is a current-quarter business reality that is scaling faster than most enterprise marketing organizations are prepared to respond to. Matt Britton regularly tells boardrooms on his keynote stages that the companies still debating whether to build for AI agents in 2027 have already missed the window that matters, which is right now, before the next major shopping season locks in agent habits for millions of households.

Industry adoption is following a similar curve on the enterprise side. Walmart has already positioned itself as the aggregator of choice across multiple agent ecosystems. Walmart already sells through Meta's Muse, ChatGPT, Gemini and other AI apps. That multi-platform posture means a brand selling on Walmart's marketplace is automatically exposed to agent traffic from several major AI assistants simultaneously, a distribution advantage Amazon's sellers currently do not have.

Brand Marketing to AI Agents: A New Playbook for CMOs

If agents are now making purchase decisions, brand marketing has to speak a language machines can parse, not just one humans find persuasive. Britton argues this requires a fundamental rebuild of how product data, reviews and pricing signals get structured and published. An agent cannot be charmed by a clever tagline, but it can be steered by clean, structured, verifiable product information.

Practically, this means marketing and ecommerce teams need to treat product feeds, structured data markup and review authenticity as core brand assets, not technical afterthoughts owned solely by IT. Walmart's gateway can help an agent find a shopper's usual items, like a preferred brand of milk, without handing over the customer's full shopping history. That means repeat-purchase categories, from grocery to household staples to beauty replenishment, are the first battleground where agent-driven loyalty will be decided, and it will be decided by data quality more than by advertising spend.

Finance and real estate brands face a parallel version of this challenge, even without a literal shopping cart. Matt Britton's work with clients across financial services and real estate shows that AI agents are already being used to compare rates, summarize disclosures and shortlist properties before a human ever speaks to a salesperson. The same trust-and-visibility logic applies: if an agent cannot cleanly parse your rate sheet or your listing data, you simply will not make the shortlist.

Generational timing makes this even more urgent. Britton's research into younger consumers, detailed across his Gen Z keynote work, shows this cohort is the least likely to manually compare products and the most likely to delegate the decision entirely to an assistant. Brands that win Gen Z loyalty over the next five years will do so largely through agent-mediated discovery rather than traditional advertising funnels, making the agent layer a generational priority, not a niche technical project.

Key Takeaways for Business Leaders

• Audit your product data feeds now for machine readability, since AI shopping agents retail decisions on structured data, not creative messaging.
• Decide your platform strategy deliberately, choosing to open your storefront to agents like Walmart or restrict access like Amazon, rather than drifting into a default.
• Monitor how your brand appears inside agent recommendation results the same way you track search rankings today.
• Build pricing governance that accounts for agent-driven price discrimination risk before regulators or watchdog groups force the issue.
• Prioritize repeat-purchase and subscription categories first, since these are the earliest and highest-volume targets for agentic commerce adoption.

Frequently Asked Questions

What are AI shopping agents in retail?

AI shopping agents are software assistants, such as Meta's Muse or ChatGPT's shopping tools, that browse products, compare prices and complete purchases on a consumer's behalf. They connect to retailer systems through protocols like MCP, allowing them to search inventory, build carts and check out without a human manually clicking through a website.

Why did Walmart allow AI agents while Amazon blocked them?

Walmart sees agent access as a distribution advantage and built Agent Profiles and the Agent Gateway to control spending limits and data access while welcoming outside agents. Amazon blocked third-party agents citing lack of authorization to access customer accounts, scrape data, or process transactions, protecting its advertising and recommendation revenue in the process.

What is the Walmart Agent Gateway?

Walmart's Agent Gateway runs on the Model Context Protocol, letting AI agents search, build carts and check out without browser automation. It is currently in limited beta for developers and companies, with Agent Profiles for individual spending limits rolling out separately within a month of the October announcement.

How should brands market to AI shopping agents instead of humans?

Brands should prioritize clean, structured product data, verified reviews and transparent pricing, since agents make selections based on machine-readable signals rather than emotional branding. Companies should also monitor how they rank inside agent recommendation outputs, treating this as a new, equally important discovery channel alongside traditional search engine optimization.

The Window to Act Is Already Closing

Walmart and Amazon just gave every retail and consumer brand executive a live, real-money experiment in how agentic commerce plays out. One retailer is betting that trust, visibility and controlled access win the next decade of AI shopping agents retail competition. The other is betting that walling off the agent layer protects a business model built on human attention, at least for now.

Matt Britton has spent his career translating fast-moving consumer shifts into action plans boards can actually execute, and this split is exactly the kind of inflection point he built his practice to address. Organizations that want a clear-eyed, data-backed briefing on what agentic commerce means for their category should explore Matt Britton's AI keynote presentations or book him directly through Speaker HQ. For ongoing analysis of how these trends unfold week to week, Britton's The Speed of Culture podcast remains one of the sharpest sources on the market.

The holiday shopping season will not wait for brands to catch up, and neither will the agents already negotiating on behalf of millions of consumers today.