Shopify lets shoppers send AI agents to buy from its catalog

Shopify is opening its catalog of over a billion listings to more AI agents, so shoppers can delegate product searches and purchases. Customers connect an agent through Shop, set shopping rules, and approve each purchase before checkout.

WTF Index NEUTRAL
◄ Terminator 1 Idiocracy 1 ►

AI agents can browse and help complete purchases, but shoppers set rules and approve every order, so the story is a routine product update with only mild implications on either scale.

Shopify lets shoppers send AI agents to buy from its catalog

Shopify is extending AI-assisted shopping beyond a single partner integration. Shoppers will be able to connect agents of their choice to its catalog, giving those agents a way to browse products and help complete purchases on a customer’s behalf.

How agent shopping works

The catalog includes over a billion listings, covering items sold both on and off Shopify storefronts. To use the feature, customers connect an agent to their account on Shop, Shopify’s platform for consolidating orders, tracking packages, and recommending products.

When a shopper chooses an item, checkout happens in the conversation with the agent through Shop Pay, Shopify’s one-click payment feature. Orders placed this way return to the Shop app, where customers can keep track of them alongside their other purchases.

Shopify describes the setup as a kind of personal shopper. Customers can give an agent instructions such as staying within a budget or showing only products with fast shipping. Those preferences guide the agent’s search, while the shopper retains a final say: each purchase requires approval.

From a single partnership to more agents

In September, Shopify announced a partnership with Meta that let Meta’s Muse agent browse and buy items for a person. The newly announced option broadens access: shoppers can use any agent without needing a partnership integration of the kind Shopify established with Muse.

The difference is meaningful for the shopping experience. Instead of starting and ending the process in a retailer’s storefront, a customer can ask an agent to find suitable products, apply personal rules, and present options for review. The agent can help with discovery and checkout, but the buyer still authorizes the transaction.

Why commerce is a focus for AI agents

Shopping is a natural demonstration for tools designed to act on a user’s behalf: the task can move from a request to a concrete transaction. The article points to examples ranging from buying clothing to booking a flight or ordering takeout. In each case, the promise is that an agent can handle steps a person would otherwise complete themselves.

That prospect has drawn investment from major technology and commerce companies. Google and Amazon are among the players building AI-driven shopping tools with the goal of having an agent make purchases for a human user. Shopify’s move places its broad product catalog within reach of more of those agent experiences.

Convenience with a human checkpoint

Delegating product search may save shoppers from comparing listings one by one, especially when they have clear limits on price or delivery. Rules such as budget guardrails and fast-shipping preferences give the agent boundaries to work within, while requiring approval for each purchase puts a person back in the decision at the moment money is spent.

That human checkpoint is central to how Shopify presents the feature. The agent can search and prepare a transaction, but it does not receive open-ended authority to buy without the shopper’s consent. Products and orders also connect back to Shop, linking the agent-assisted purchase to the platform customers use to follow their shopping activity.

Whether this approach becomes a routine way to shop will depend on how useful shoppers find agents for choosing products and completing transactions. Shopify is making the underlying catalog available to a wider range of agents, while its Shop account connection, Shop Pay checkout, and purchase approval define the path from recommendation to order.