🔍 Read the full analysis: Walmart Welcomes AI Shopping Agents. What Does Amazon’s Block Mean? on ThorstenMeyerAI.com
Prime made for students and young adults
- Fast, free delivery for dorm and study essentials
- Prime Video and Amazon Music included
- Member-only deals
TL;DR
Walmart announced an Agent Gateway and customer-controlled Agent Profiles on Oct. 21, opening a structured route for approved AI agents to search and buy. Amazon has blocked Meta’s Muse agent, arguing that it violates its terms and raises security concerns; the broader dispute is over access, customer control and who receives credit for sales.
Walmart announced an Agent Gateway and Agent Profiles on Oct. 21, creating a structured way for approved AI shopping agents to search its offerings and make purchases under customer-set permissions. The move contrasts with Amazon’s block on Meta’s Muse, sharpening a dispute over which agents retailers will admit, how shoppers control automated buying and who gets credit for a sale.
Walmart’s gateway is based on the Model Context Protocol (MCP), a technical framework that lets software agents interact with services through structured tools rather than navigating a website as a person would. Walmart also announced Agent Profiles, which the source describes as a way for customers to set spending limits and permissions. The announcement does not establish how broadly the system is available or which agents can use it now.
Meta launched its Muse personal agent on Sept. 8. The company says Muse can use a browser, fill out forms and act for a user. On Sept. 20, users reportedly encountered an Amazon message 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’s launch post said Muse had no visibility into users’ passwords or payment methods.
Meta reported on its Oct. 21 earnings call that more than 10 million people in the United States and Canada had set up Muse in roughly six weeks. Meta said shopping was its most common use and the average purchase exceeded $200. Those figures are company-reported; the source does not provide an independent measurement or further detail about how purchases were counted.
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 Set the Rules for Agents
The disagreement affects whether shoppers can use one assistant across different stores or must use each retailer’s own tools. A browser-based agent can interact with sites much like a human shopper, but retailers may have less visibility into what it is doing and how it handles credentials. A retailer-built gateway offers more structured access, while giving the retailer greater control over what the agent can see and do.
There is also a commercial question: who owns the customer relationship when an agent researches products and completes an order? Meta chief executive Mark Zuckerberg said the company expected to profit by taking a small fee from transactions. Amazon generated more than $68 billion in advertising revenue last year, according to GeekWire, making control over shopping discovery and purchasing relevant to its business as well as site security.
Publishers and affiliate businesses could also be affected. The source notes that Amazon Associates commissions are tied to tracked links. If an agent reads a review but purchases through another route, that link may not receive credit. The scale of any such effect is not established in the material provided.
Popular right now
Samsung 42-Inch S90H OLED 4K Smart TV Bundle
Samsung 42-Inch Class OLED S90F 4K Smart TV (2025)
Bose QuietComfort Wireless Noise-Cancelling Over-Ear Headphones, Black
As an affiliate, we earn on qualifying purchases.
Three Routes Into Online Shopping
The dispute reflects three ways an AI agent can shop. It can use a browser as a person would, opening pages and interacting with forms. It can connect through a retailer’s designated tools, such as Walmart’s announced gateway or Shopify’s checkout tools. Or it can use shared protocols intended to describe how agents, merchants and payment services communicate.
Several protocol efforts are under way. The source describes the Personal Agent Protocol, or Poppy, backed by Meta, Sierra, Walmart, Shopify and Stripe among others, as a draft focused on agent identity and customer permission. It also describes Google and Shopify’s Universal Commerce Protocol and OpenAI and Stripe’s Agentic Commerce Protocol. These projects have different participants and functions; their existence does not mean retailers have adopted a common standard.
Earlier developments show that implementation remains unsettled. Meta announced retail connectors in September, naming Walmart and several other retailers. Shopify introduced structured checkout tools for browser agents. OpenAI had launched Instant Checkout in ChatGPT using its protocol in 2025, but the source says it stepped back from that initial approach in March 2026, citing a need for greater flexibility and a focus on product discovery.
Access, Safety and Adoption
The source does not say when Walmart’s gateway will be available to shoppers, which outside agents are approved, or how Agent Profiles will be enforced in practice. It is also unclear whether Walmart’s system will handle purchases directly across its full assortment or initially cover a narrower set of products and transactions.
The central security dispute remains unresolved in the material provided. Amazon said Muse did not identify itself and appeared to store credentials; Meta said Muse could not see passwords or payment methods. No independent technical assessment is provided to settle those competing accounts. The source also says Muse encountered Target’s bot checks inconsistently, while users had difficulty with services including Delta and Amtrak, underscoring that access varies by site.
Meta’s adoption and spending figures are company claims, not independently verified results in the source. It is also too early to know how often agents complete purchases, how much retailers will charge or pay for agent access, and whether shared protocols will gain broad adoption.
Rollout and Protocol Decisions
The next developments to watch are Walmart’s rollout details: availability, participating agents, customer permission settings and the purchase categories supported. Retailers and agent providers will also need to clarify how identity, payment authorization, returns and customer support work when an automated assistant places an order.
Protocol adoption will be another test. The source describes Poppy as an early draft, while other commerce protocols have different backers and functions. Further retailer announcements, published technical rules and independent reporting on how agents handle credentials and transactions should help show whether the industry moves toward shared access standards or continues with retailer-by-retailer controls.
For shoppers, the practical advice in the source is to check an agent’s spending permissions before use. Purchase approval, spending caps and payment controls can limit unintended orders. Buyers should also confirm which retailer’s return policy applies and who to contact if an agent makes a mistake.
Key Questions
What did Walmart announce?
Walmart announced an Agent Gateway based on MCP and customer Agent Profiles. The gateway is intended to let approved agents access structured shopping tools; the source does not specify a full public rollout date.
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 said Muse had no visibility into passwords or payment methods. The source provides no independent finding that resolves the disagreement.
Can AI agents already complete purchases?
Some agent and checkout tools have been introduced, but access depends on the retailer and the system. The source describes browser-based shopping, retailer gateways and shared protocols; it does not establish that agents can reliably buy across online stores.
How might AI shopping agents affect affiliate links?
Affiliate commissions can depend on a shopper using a tracked link. If an agent researches a product and completes the order through a different route, the publisher that informed the shopper may not receive commission. The size of that effect is not known.
Source: ThorstenMeyerAI.com
