And while it can be viewed, in one sense, as an extension of the traditional retail context for proximity — first physical, then digital and now increasingly cognitive — the emerging battleground is being shaped by the impact of artificial intelligence on customer interfaces.

Amazon and Walmart, for example, are already racing to move beyond the limits of who has the best products or prices to compete over who can best understand and respond to the choices consumers make.

The new end zone for retail innovation is being fought across the “decision layer” of AI commerce. The emerging retail decision layer is not a single product or feature. It is an evolving system of recommendation engines, conversational agents, predictive pricing and fulfillment orchestration that collectively determines what consumers see, consider and ultimately buy.

In this new paradigm, winning is less about stocking shelves or optimizing search results, and more about owning the emergent layer that sits between intent and purchase.

For Walmart and Amazon, the stakes are existential. If third-party AI assistants or platform-agnostic agents become a primary customer interface for shopping, the retailer could risk becoming a commoditized infrastructure layer.

Read also: Amazon and Walmart Want More Than Your Prescription

The Rise of the Decision Layer

Historically, retail competition revolved around assortment, price and convenience. Online platforms like Amazon shifted the balance toward search and discovery, using algorithms to guide users through vast inventories. But the rise of generative AI and conversational interfaces is compressing that journey. Instead of browsing pages of results, consumers increasingly expect a single, context-aware recommendation — or even a fully executed purchase — based on minimal input.

PYMNTS Intelligence’s Prompt Economy work shows how far along this shift already is. Nearly 70% of consumers say they are interested in using artificial intelligence agents to simplify shopping tasks; more than half would like an autonomous agent to monitor and do their weekly shopping for them, or look through personal interactions with a friend to identify and purchase a gift.

Walmart’s recent push into AI can be seen as representing a calculated effort to leapfrog traditional eCommerce limitations and redefine its role in the customer journey. The company has been experimenting with generative AI tools that allow users to describe needs in natural language, such as “plan a week of healthy meals under $100,” and receive curated shopping lists that can be purchased in a single click.

Crucially, Walmart’s advantage lies in its hybrid model. With thousands of physical stores functioning as fulfillment hubs, the company can integrate real-time inventory data into its AI systems, enabling more accurate recommendations and faster delivery. This tight coupling of digital intelligence and physical infrastructure may allow Walmart to compete not just on price, but on certainty, knowing that what is recommended is available and can be delivered quickly.

Read more: Mid-Tier Retailers Caught Between Amazon and Walmart

Amazon, by contrast, is extending its long-standing dominance in algorithmic retail into a more autonomous future. The company has been layering generative AI capabilities into its platform, from enhanced product summaries to conversational shopping assistants that can guide users through complex decisions.

But Amazon’s real strength lies in its end-to-end control of the commerce stack. From product discovery to payment to last-mile delivery, Amazon has built a vertically integrated system that can be optimized holistically. This allows it to experiment with machine-led commerce models where the system not only recommends products but also dynamically adjusts pricing, bundles items, and schedules delivery based on predicted demand and user behavior.

Amazon’s investments in logistics, including same-day delivery and automated warehouses, feed into this system, enabling faster and more reliable fulfillment that reinforces customer trust in machine-generated decisions.

See also: Small Businesses Stop Chasing Amazon on Delivery Speed

Fulfillment as a Determinant of Choice

In the context of the decision layer, fulfillment is not an afterthought but a key input. AI systems must account for delivery times, shipping costs and inventory constraints when generating recommendations. A product that is slightly more expensive but available for immediate delivery may be favored over a cheaper alternative with a longer lead time.

Similarly, as the decision layer evolves, pricing is no longer a static attribute but a dynamic signal that can influence and be influenced by AI systems. The system must balance cost, quality, availability and delivery speed to arrive at the “best” option for the user. This requires integrating pricing data with other variables in a way that is transparent enough to build trust, yet sophisticated enough to maximize margins.

For consumers, the shift happening across retail may promise greater convenience and personalization. For retailers, it can raise new questions about control, differentiation, and trust. And for Walmart and Amazon, it could represent the next frontier in a long-running competition — one that will be decided not just by which company has the best products or prices, but by which one can best understand and shape the choices consumers make.