🔍 Read the full analysis: How Amazon And Walmart Are Handling AI Shopping Agents Differently on ThorstenMeyerAI.com
Get hardware and tech essentials delivered free with Prime
- Fast, free delivery on millions of items
- Prime Video, Amazon Music and more included
- Member-only deals all year
TL;DR
Walmart announced an Agent Gateway and customer-controlled Agent Profiles, while Amazon has blocked Meta’s Muse and says outside agents must identify themselves. The split reflects a wider dispute over how agents access stores, who controls checkout and customer relationships, and how purchases are attributed.
Walmart is opening a structured route for AI shopping agents, while Amazon has blocked Meta’s Muse and says outside agents must identify themselves under its Conditions of Use. The contrasting policies, reported on Oct. 21, put retailers at the center of a growing dispute over how automated assistants browse and buy products on behalf of shoppers.
Walmart announced Agent Profiles and a Walmart Agent Gateway, which uses the Model Context Protocol, or MCP, to let approved agents search inventory and make purchases through a structured connection. The company’s profiles are intended to let customers set permissions, including spending limits. The announcement does not mean every agent can access Walmart: the gateway is a designated route, and details about availability and approval requirements remain limited in the source material.
Amazon has taken a more restrictive position toward Meta’s Muse, which uses a browser in a virtual computer hosted in Meta’s cloud. Amazon cut off Muse in September. The company said the agent had not identified itself, that Meta had not informed Amazon it would shop on the site, and that Muse appeared to capture and store customer credentials. Meta’s launch post said Muse “has no visibility into people’s passwords or payment methods.” Those are opposing statements from the companies, not a resolved finding about how Muse handled credentials.
Meta says Muse reached more than 10 million users in the U.S. and Canada within about six weeks of launch, with shopping its most common use and an average purchase above $200. Those figures came from Meta’s Oct. 21 earnings call and have not been independently verified in the source material. The adoption claims help explain why retailers are setting policies now: an agent that can search, compare and place orders may become another route between shoppers and stores.
Agentic shopping
Three ways an AI agent buys for you
Every fight between shops and AI agents comes down to which of these three routes the agent uses, and whether the shop agreed to it.
The browser, like a human
The agent opens pages built for people, signs in with your login and clicks through checkout. Works almost anywhere, which is why shops object: they can't tell the agent from you.
Examples: Meta's Muse (from a virtual machine in Meta's cloud), Perplexity's Comet (from your own computer)
The shop's own agent door
Approved agents search and buy through a structured interface instead of clicking through pages. The shop sets the rules.
Examples: Walmart Agent Gateway (MCP, beta), Shopify MCP server and WebMCP checkout tools
Shared protocols
Common rules for who the agent is, what you allowed, and how the checkout and payment are confirmed.
Examples: Personal Agent Protocol (draft), Universal Commerce Protocol, Agentic Commerce Protocol
The protocols competing to be the standard
| Protocol | Backers | Covers | Status |
|---|---|---|---|
| Personal Agent Protocol (Poppy) | Sierra, Meta, Walmart, Shopify, Stripe + 35 design partners | Agent identity, customer permission, one session across site, API and company agent | Draft 0.1 (9 Oct 2026) |
| Universal Commerce Protocol (UCP) | Google, Shopify; site also lists Etsy, Wayfair, Target, Walmart, Amazon, Microsoft | Discovery, cart, checkout, orders; payment mandates via AP2 | Published |
| Agentic Commerce Protocol (ACP) | OpenAI, Stripe | Checkout between agent and merchant | Open source; ChatGPT's own checkout scaled back in March 2026 |
What a careful agent purchase looks like
- You give the task and a limit, for example a $50 cap.
- The agent identifies itself to the shop and gets only the access you approved.
- It prepares the cart and shows you an approval card.
- Payment goes through a token or one-time card, so the agent never sees your card.
- The shop’s normal returns apply if the agent got it wrong.
Sources: Meta (8 Sep 2026), Walmart (21 Oct 2026), TechCrunch on Shopify (28 Sep 2026), Sierra (9 Oct 2026), ucp.dev, agenticcommerce.dev, OpenAI (24 Mar 2026). Purchase steps: Walmart’s agent terms and Meta’s Muse description.
Powered by Thorsten Meyer AI
Control of the Shopping Relationship
The divide is not simply whether retailers support AI. It concerns who controls the shopping process: the retailer, the agent company or the customer. A browser-based agent can act through a shopper’s account on pages designed for people. That may work across many sites, but retailers say they need to distinguish automated activity, protect accounts and set terms for access. A structured gateway gives a store more control over which agents connect and what they can do.
There is also a commercial question. If an agent recommends a product and completes a purchase without a conventional referral click, retailers and publishers may have difficulty assigning credit for the sale. The source reports that Amazon’s Associates program pays commissions tied to tracked links, and that Meta has said it expects to earn “a small fee from transactions.” How fees, referral attribution and advertising will work across agent-led purchases is not settled by Walmart’s announcement.
For shoppers, convenience can come with practical risks. An agent may spend money, select the wrong product or place an order under a retailer’s return rules. Walmart’s profile controls point toward one possible safeguard, but the source does not establish what protections are available across all agents or stores. Customers should check permissions and payment arrangements before allowing an assistant to buy on their behalf.
Products Worth Considering
Apple MacBook Air 15-inch Laptop with M5 Chip, 16GB Memory, 512GB SSD, and Liquid Retina Display
Ledger Nano Gen5 Crypto Wallet, Black
LG UltraGear 27-Inch QHD OLED Gaming Monitor, 240Hz, 0.03ms, G-SYNC Compatible, FreeSync Premium Pro, HDMI 2.1
As an affiliate, we earn on qualifying purchases.
From Browser Bots to Retail Gateways
The recent developments have moved quickly. Meta launched Muse on Sept. 8, describing an agent that can open a browser, fill in forms and act for a user. On Sept. 20, Amazon blocked Muse. Three days later, Meta announced retail connectors involving Walmart, Best Buy, Gap, Sephora, Wayfair and Expedia. Shopify has also developed structured tools that let browser agents interact with checkout, rather than relying only on screen-based clicking.
On Oct. 6 and 9, Meta and Sierra announced the Personal Agent Protocol, also called Poppy, with Walmart, Shopify and Stripe, then published a draft with additional design partners. Poppy is one of several efforts to define how agents identify themselves and carry user permissions between services. The source describes it as a draft and says payments are left to extensions; it should not be treated as a settled industry standard.
The source also identifies competing commerce protocols, including the Universal Commerce Protocol associated with Google and Shopify, and the Agentic Commerce Protocol from OpenAI and Stripe. Their existence shows that businesses are testing different ways to connect agents with merchants. Protocols and partnerships alone do not establish that shoppers can use the same agent to complete purchases reliably across retailers.
“Continued access by an unauthorized AI agent violates Amazon’s Conditions of Use, to which our customers have agreed.”
— Amazon, in the message shown to Muse users
Access Rules and Purchase Safeguards
Several details remain open. The source does not specify which agents Walmart will approve, when its gateway and profiles will be available to customers, or exactly what limits and review steps the profiles will support. Amazon’s stated concerns about Muse are disputed by Meta, and the source does not include independent technical findings about the agent’s handling of credentials.
It is also unclear how widely agents can complete purchases rather than simply research products. Data cited from HUMAN Security says 81% of agent requests went to product and search pages and 2.4% to checkout and payment in its September data. The source does not give a fuller account of the sample or measurement method, so those figures should not be generalized to every agent or retailer. The payment, returns, liability and referral-credit rules for purchases made through different systems are still developing.
Retailer Rollouts and Protocol Tests
The immediate next step is whether Walmart turns its announcement into a working service: which agents can connect, what customers can authorize and how purchases are confirmed. Retailers and agent providers are also likely to keep testing browser access alongside structured connections such as MCP, while the competing protocol groups refine how identity, permission and checkout work together.
For customers, the practical test will be whether an agent can find an item, show its source and price, ask for approval at the right point, and complete an order with clear payment and return terms. The source material does not provide a timetable for broad availability or a common set of rules. Until those details emerge, Amazon’s restrictions and Walmart’s gateway represent different approaches rather than a settled model for AI shopping.
Key Questions
What is Walmart’s Agent Gateway?
It is a structured connection for approved AI agents to search Walmart inventory and make purchases. Walmart also announced Agent Profiles intended to let customers set permissions such as spending limits.
Why did Amazon block Meta’s Muse?
Amazon said Muse had not identified itself, that Meta had not told Amazon it would shop on the site, and that the agent appeared to capture and store customer credentials. Meta said Muse had no visibility into passwords or payment methods. The source does not independently resolve the disagreement.
Can AI shopping agents already complete purchases?
Some systems are designed to support checkout, but the source does not establish that agents can reliably buy across retailers. The cited September data says most measured agent requests went to product and search pages, while a smaller share went to checkout and payment.
What should shoppers check before letting an agent buy?
Review the agent’s permissions, spending limits, payment method and the retailer’s return terms. These controls matter because an agent may place an order on the customer’s behalf, and the source does not describe uniform safeguards across services.
Source: ThorstenMeyerAI.com
