Amazon has blocked Muse, Meta’s new artificial intelligence assistant, from shopping on the user’s behalf.
The online retail giant said on Monday that it had not been notified by Meta in advance that Muse would access its online store, and that Amazon had not authorized the AI agent to do so.
“We think it’s fairly straightforward that third-party applications that offer to make purchases on behalf of customers from other businesses should operate openly and respect service provider decisions about whether or not to participate,” a spokesperson said in a statement to The Epoch Times.
The spokesperson compared the arrangement to food-delivery apps working with restaurants or online travel agencies booking plane tickets with airlines.
“Agentic third-party applications such as Muse have the same obligations, and we’ve requested that Meta remove Amazon from the experience.”
Meta introduced Muse on Sept. 8 as a personal AI agent designed to perform tasks on a user’s behalf rather than simply answer questions.
According to Meta, the AI agent can use its own browser to fill out forms, send emails, book travel, and shop online, among other tasks.
Beginning on the night of Sept. 20, people who tried to use Muse to shop on Amazon were met with a message reading: “Continued access by an unauthorized AI agent violates Amazon’s Conditions of Use, to which our customers have agreed.”
Privacy and Security Concerns
According to Amazon, Muse does not identify itself when accessing the retailer’s website and appears to capture and store customer credentials while accessing account information.
The company said the practice raises questions about how sensitive customer information is handled when an outside AI agent accesses an Amazon account.
Meta did not respond to a request for comment, but the company has described Muse’s behavior differently.
When announcing Muse, Meta said the agent operates inside a secure, dedicated virtual machine, and that users determine which services it connects to and how much access it can have.
Meta also says Muse cannot see users’ passwords or payment information. Credentials are instead kept in secure storage that allows the agent to use them without viewing the information directly.
Muse will ask for user approval before taking “sensitive actions” such as sending an email or making a purchase, according to the company.
Amazon and Meta have cooperated in other areas of online shopping. Since 2023, customers have been able to shop through Amazon’s promotion ads on Facebook and Instagram and check out with Amazon without having to leave the social media sites.
It’s not clear whether the Muse dispute will affect those partnerships.
Battle Over AI Shopping
This is not the first time Amazon sought to restrict other companies’ AI shopping agents from accessing its marketplace.
Last November, Amazon sued Perplexity over Comet, a web browser with a built-in AI agent capable of logging into users’ shopping accounts and placing orders on their behalf. Amazon accused the startup of trying to “conceal” its AI agents so they could continue accessing the retailer’s website without permission.
A federal judge initially sided with Amazon and temporarily blocked Perplexity’s shopping agent. But on Aug. 4, a three-judge panel of the Ninth U.S. Circuit Court of Appeals overturned the injunction, concluding that Amazon was unlikely to succeed on its claim that Perplexity violated the federal Computer Fraud and Abuse Act.
The judges found that, for purposes of that law, it was Amazon customers using Perplexity’s AI tools who accessed Amazon’s systems, rather than Perplexity itself. The ruling did not resolve Amazon’s broader lawsuit against the startup.
Amazon, meanwhile, operates its own AI-powered shopping service called Buy for Me. The feature allows customers to purchase certain products from third-party retailers when those items are not sold directly through Amazon.
Amazon says merchants that do not want their products included in the service can opt out.

