🔍 Read the full analysis: Amazon Blocks AI Shopping Agents. Walmart Lets Them In. Now What? on ThorstenMeyerAI.com
Get smart everyday buys 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
Amazon blocked Meta’s Muse from accessing its shopping site, while Walmart announced an Agent Gateway and profiles that let approved agents search and buy under customer-set limits. The competing approaches reflect a broader dispute over who controls shopping data, checkout and the customer relationship; how widely the new tools will work remains uncertain.
Walmart is opening a route for approved AI shopping agents through its Agent Gateway and customer-set Agent Profiles, while Amazon has blocked Meta’s Muse from accessing its store. The split, reported in October 2026, puts retailers’ control over automated shopping in focus as Meta says millions of people have tried its agent.
Walmart announced the Gateway and Profiles on October 21. The source report says the Gateway uses the Model Context Protocol, or MCP, to let approved agents search inventory and place orders through a structured connection rather than navigating the store as a person would. Agent Profiles are intended to let customers set permissions, including spending limits. The announcement does not establish how broadly the tools are available to shoppers or which agents can use them now.
Amazon took the opposite approach with Muse. On September 20, users attempting to use the agent on Amazon encountered a notice saying continued access by an unauthorized AI agent violated the retailer’s Conditions of Use. Amazon said Muse had not identified itself as an agent and appeared to capture and store customer credentials. Meta disputed the password claim, saying in its launch material that Muse had no visibility into people’s passwords or payment methods. The dispute has not been independently resolved in the supplied material.
Meta said on its October 21 earnings call that more than 10 million people in the United States and Canada had set up Muse in roughly six weeks, and that shopping was its most common use. Meta also reported an average purchase above $200. Those figures are the company’s own metrics, not an independently audited measure of completed agent purchases. Meta has said it expects to earn a small fee from transactions.
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
Retailers Compete for Agent Access
The disagreement is about more than whether software can browse a product page. A shopping agent can shape what a customer sees, compare prices, select a retailer and potentially complete a purchase. Retailers want to decide how that agent enters, what information it can retrieve and which checkout rules apply. Opening a structured connection gives a retailer more control than allowing an automated browser to act through a customer’s account.
There are also commercial stakes for platforms and publishers. The source report cites GeekWire’s figure of more than $68 billion in Amazon advertising revenue last year and says Meta wants a share of the value Muse creates. If an agent becomes the shopper’s main interface, the retailer could lose influence over discovery and advertising, while the agent operator may seek a transaction fee. Affiliate publishers may also miss commissions when an agent reads a review but purchases without following a tracked link.
For shoppers, the convenience comes with questions about authorization and responsibility. A spending cap or approval requirement can limit unwanted orders, but consumers still need to know which account and payment method an agent uses, and which retailer’s return policy applies. Access does not by itself settle who is accountable if an agent selects the wrong item or makes a purchase the customer did not intend.
Products Worth Considering
FLEXISPOT 42-Inch Electric Height-Adjustable Desk Converter with Keyboard Tray
Brother TD-2020A 2‑Inch Direct Thermal Desktop Label Printer with USB
Apple iPad Air (2025) 13‑inch Wi‑Fi 256 GB – Space Gray (M3 chip)
As an affiliate, we earn on qualifying purchases.
From Browser Bots to Store Gateways
The dispute follows a short series of developments. Meta launched Muse on September 8, describing it as an agent that can open a browser, fill in forms and act for a user. The report says Muse runs in a dedicated virtual machine in Meta’s cloud and uses Link by Stripe for one-time-use cards. On September 23, Meta announced retail connectors involving Walmart and several other companies. Walmart’s October announcement followed as retailers and technology firms began building more structured ways for agents to interact with stores.
These systems use different routes. A browser-based agent visits pages built for people and may use the customer’s login; a store gateway exposes approved functions directly; and shared protocols seek to standardize what an agent can request or do. The source report describes the Personal Agent Protocol, backed by Meta, Sierra and partners including Walmart and Shopify, as a draft, while Google and Shopify’s Universal Commerce Protocol and OpenAI and Stripe’s Agentic Commerce Protocol take different approaches. The protocols are not interchangeable, and their existence does not mean retailers support every agent or that purchases will work consistently.
Early usage data also suggests that agent activity is not the same as agent checkout. The report attributes to HUMAN Security’s September data a finding that 81% of agent requests went to product and search pages, while 2.4% went to checkout and payment. The figures describe requests in that dataset and period; they do not establish the share of all online shopping or predict future purchases.
How Much Agent Shopping Works
It remains unclear how many Muse users have completed purchases through the agent, how often customers approve transactions, and how Walmart’s Gateway will perform once available to a wider group of agents. Meta’s setup and purchase figures are company-reported and do not specify, in the source material, the number of completed transactions or the period covered by the average purchase figure beyond the roughly six-week launch window.
The source report also does not establish whether Amazon will offer a structured connection for outside agents, whether Walmart will permit all agents or only approved partners, or how the companies will handle disputes, returns and errors when an agent initiates an order. Amazon’s allegations about Muse and Meta’s denial remain competing statements; the material does not provide an independent technical finding. Retailer policies and protocol support may change as the systems develop.
Retailer Access Will Set the Pace
The next test is practical availability: which agents can connect to Walmart’s Gateway, what customers can authorize through Agent Profiles, and whether the process supports reliable checkout as well as product search. Retailers and protocol developers are also likely to refine identity, consent and payment rules as they test integrations. The source report describes several standards as published or in draft, but gives no common launch date or shared adoption schedule.
For shoppers, the immediate approach is to check an agent’s permissions before connecting it to a store account. Use purchase approvals and spending limits where available, review the payment method, and confirm the retailer’s return terms. Whether Amazon changes its policy or Walmart’s open route becomes widely used is still unknown; the outcome will depend on how retailers balance customer convenience with control over their stores and transactions.
Key Questions
Why did Amazon block Meta’s Muse?
Amazon said Muse was an unauthorized agent that did not identify itself and appeared to capture and store customer credentials. Meta disputed the credential concern, saying Muse could not see users’ passwords or payment methods. The supplied material does not include an independent resolution of the disagreement.
What is Walmart’s Agent Gateway?
It is a structured connection announced by Walmart that uses MCP to let approved agents search inventory and buy without simply clicking through the retailer’s consumer-facing pages. Walmart also announced Agent Profiles with customer permissions and spending limits. Availability and partner access are not fully specified in the source material.
Does Meta’s user figure mean millions have bought through Muse?
No. Meta said more than 10 million people in the United States and Canada had set up Muse within about six weeks, and that shopping was its most common use. The figure does not say how many people completed a purchase. It is a company-reported setup count, not a confirmed transaction total.
What should shoppers check before using an AI shopping agent?
Review what the agent can access, set a spending limit or approval requirement, and check which payment method it will use. Shoppers should also confirm which retailer’s return policy applies and how to challenge an order if the agent makes an error.
Source: ThorstenMeyerAI.com
