AIThis post was created with the assistance of artificial intelligence (AI).

🔍 Read the full analysis: Amazon And Walmart Take Opposite Sides On AI Shopping Agents on ThorstenMeyerAI.com

Buying for a business?Offer from Amazon

Get business pricing on tech for your team

  • Business-only prices and quantity discounts
  • Tax-exempt purchasing
  • Multiple users, one account, clear invoices
As an affiliate, we earn on qualifying purchases.

TL;DR

Walmart announced tools designed to let approved AI shopping agents access its product catalog and make purchases, while Amazon has blocked Meta’s Muse agent from its site. The split puts retailers’ control over customer access, checkout and transaction data at the center of a developing contest.

Walmart is opening a route for AI shopping agents to search its inventory and make purchases, while Amazon has blocked Meta’s Muse from accessing its site. The opposing approaches, outlined in developments reported in October, mark a growing dispute over whether outside agents can shop on retailers’ behalf—and who controls the customer relationship when they do.

Walmart announced Agent Profiles and a Walmart Agent Gateway on October 21. The gateway is designed to let approved agents use structured tools to search Walmart’s inventory and buy products rather than navigating the retailer’s website as a human would. Profiles are intended to give an agent permissions and spending limits. The announcement describes Walmart’s approach; it does not establish how widely the tools are already available or which agents can use them.

Amazon has taken the opposite position with Meta’s Muse. After Muse was blocked from Amazon on September 20, Amazon said the agent had not identified itself and appeared to capture and store customer credentials. Meta’s launch post said Muse “has no visibility into people’s passwords or payment methods.” The companies’ accounts differ, and the available information does not independently resolve the dispute.

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. Meta also said shopping was its most common use and the average purchase was over $200. These are company-reported figures, not independent measurements of completed purchases or repeat usage.

At a glance
reportWhen: Walmart announcement and Meta figures r…
The developmentWalmart has announced an agent gateway and spending profiles as Amazon bars Meta’s Muse from shopping on Amazon.
Thorsten Meyer AI
How it works

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.

1

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)

2

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

3

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

ProtocolBackersCoversStatus
Personal Agent Protocol (Poppy)Sierra, Meta, Walmart, Shopify, Stripe + 35 design partnersAgent identity, customer permission, one session across site, API and company agentDraft 0.1 (9 Oct 2026)
Universal Commerce Protocol (UCP)Google, Shopify; site also lists Etsy, Wayfair, Target, Walmart, Amazon, MicrosoftDiscovery, cart, checkout, orders; payment mandates via AP2Published
Agentic Commerce Protocol (ACP)OpenAI, StripeCheckout between agent and merchantOpen source; ChatGPT's own checkout scaled back in March 2026

What a careful agent purchase looks like

  1. You give the task and a limit, for example a $50 cap.
  2. The agent identifies itself to the shop and gets only the access you approved.
  3. It prepares the cart and shows you an approval card.
  4. Payment goes through a token or one-time card, so the agent never sees your card.
  5. 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

Who Controls the Agent Checkout

The split matters because an agent can change who mediates a purchase. If shoppers ask an AI to compare products and select a seller, the agent may influence what gets seen, where a shopper buys, and which retailer receives the transaction. Retailers want control over inventory access, checkout and customer data; agent providers want to make shopping easier and may seek a share of the value generated.

There are trade-offs for consumers. A retailer-approved connection could make an agent’s permissions and checkout more explicit, while a browser-based agent may work across more sites but prompt concerns about identification, account access and terms of use. Before delegating purchases, users need to know what the agent can spend, whether a purchase needs approval, and which seller’s return policy applies.

The commercial stakes reach beyond retailers and AI companies. Product reviews and affiliate links often generate revenue when a shopper clicks through to a merchant. An agent that reads a review but completes a purchase through another route may not generate that tracked click. The source material describes this as a risk to affiliate publishers, not as a measured industry-wide loss.

Products Worth Considering

DJI Osmo Mobile 8 Phone Gimbal with Mic Mini Transmitter, Infinity Black

DJI Osmo Mobile 8 Phone Gimbal with Mic Mini Transmitter, Infinity Black

Amazon Kindle Paperwhite Signature Edition (12th Generation, 2024), 32 GB, Metallic Black

Amazon Kindle Paperwhite Signature Edition (12th Generation, 2024), 32 GB, Metallic Black

Sony WH-1000XM6 Wireless Noise-Canceling Over-Ear Headphones, Blue, Bundle

Sony WH-1000XM6 Wireless Noise-Canceling Over-Ear Headphones, Blue, Bundle

As an affiliate, we earn on qualifying purchases.

From Browser Bots to Retail Gateways

Muse launched on September 8 as a personal agent that Meta said can open a browser, fill out forms and act for a user. It operates in a dedicated virtual machine in Meta’s cloud and uses Link by Stripe for payments, according to the source material. The browser approach can reach sites built for people, but it can also make it harder for a retailer to distinguish an agent from a customer.

After Amazon blocked Muse, Meta announced retail connectors with Best Buy, Gap, Sephora, Walmart, Wayfair and Expedia on September 23. Mark Zuckerberg said Meta expected to earn “a small fee from transactions.” Walmart’s October gateway represents a different route: retailers can provide structured access rather than letting an agent interact with ordinary web pages.

Other companies are developing tools and protocols for agent-led commerce, but standards remain in motion. The Personal Agent Protocol, backed by Meta, Sierra and partners including Walmart, Shopify and Stripe, had a draft available in October. Google and Shopify’s Universal Commerce Protocol and OpenAI and Stripe’s Agentic Commerce Protocol are other efforts. OpenAI said in March 2026 that it was stepping back from its initial Instant Checkout approach in ChatGPT to focus on product discovery while merchants use their own checkout experiences. These initiatives show activity, not broad adoption or an agreed standard.

“Continued access by an unauthorized AI agent violates Amazon’s Conditions of Use, to which our customers have agreed.”

— Amazon, in a message shown to Muse users

Access, Safety and Adoption Remain Open

It is not yet clear how many Muse users have completed purchases, how often they return to use the agent, or how Meta’s reported average purchase was calculated. The company’s setup figures do not by themselves show how much shopping is being completed through agents.

The dispute over Muse’s access to Amazon also remains unresolved in the supplied accounts. Amazon raised concerns about identification and credentials; Meta denied visibility into passwords and payment methods. Further technical evidence or a detailed response from either company is not provided here. The breadth of Walmart’s gateway launch, its eligibility rules and the protections available to shoppers also remain unclear.

More broadly, the presence of competing protocols does not establish which, if any, will become widely used. The source material cites HUMAN Security data from September saying 81% of observed agent requests went to product and search pages and 2.4% to checkout and payment. The sample and measurement details are not supplied here, so those figures should be read as the reported dataset’s observations, not as a measure of all AI shopping.

Retailers Set the Next Access Rules

The next test is whether Walmart’s gateway moves from an announcement into broad use by agents and shoppers, and whether other retailers adopt similarly structured access. Companies will also need to explain how agent permissions, spending caps, payment authorization, refunds and returns work in practice.

Amazon’s policy toward outside agents and Meta’s response could shape whether the dispute stays focused on Muse or becomes a wider fight over automated browsing. Protocol developers and retailers may publish further specifications, partnerships or usage figures. Until those details emerge, shoppers should treat agent checkout as a developing service and review purchase approvals, spending limits and return terms before relying on it.

Key Questions

What did Walmart announce?

Walmart announced Agent Profiles and an Agent Gateway, designed to let approved AI agents access inventory and make purchases through structured tools. The source material does not specify the full rollout or availability.

Why did Amazon block Meta’s Muse?

Amazon said Muse had not identified itself and appeared to capture and store customer credentials, and said its access violated Amazon’s Conditions of Use. Meta said Muse has no visibility into users’ passwords or payment methods. The dispute is not independently resolved in the supplied information.

How many people have set up Muse?

Meta said on October 21 that more than 10 million people in the United States and Canada had set up Muse in roughly six weeks. The figure is company-reported and does not state how many completed purchases.

Does Walmart’s announcement mean AI agents can shop everywhere?

No. The gateway is a Walmart-specific route, and retailers set their own access rules. The source material describes competing protocols and uneven access, not a universal system accepted by online stores.

What should shoppers check before using an agent?

Check whether purchases require approval, set a spending cap, understand which payment method is used, and confirm the retailer’s return policy. These details matter because an agent may act on the user’s behalf under retailer-specific rules.

Source: ThorstenMeyerAI.com

You May Also Like

What Happens When AI Builds A Stickman Game? An Arcade Takes Shape

Claude Opus 5.5 built BEATDOWN, a rhythm stickman brawler, plus nine browser games in a single day with no hand-written code.

Otter.ai Raises $320 Million From General Catalyst’s Growth Fund For Its Meeting Tool

Otter.ai says it will use $320 million from General Catalyst’s Customer Value Fund to expand enterprise sales, international reach and product development.

The Ultimate Guide To Fine-tuning 350M AI Models For Improved Output Consistency

Liquid AI releases an open-source recipe to improve structured output compliance of the LFM2.5-350M model using GRPO, boosting benchmark scores on a free GPU setup.