Just weeks after launching and quickly rising to the top of the App Store charts, Meta’s new AI assistant, Muse, has already run into trouble with Amazon.

According to GeekWire, users attempting to shop on Amazon through Muse began seeing a pop-up message on Sunday night stating that continued access by an unauthorized AI agent violates Amazon’s Conditions of Use.

Amazon later confirmed to Gizmodo that it has asked Meta to stop allowing Muse to make purchases on Amazon.com on behalf of users.

An Amazon spokesperson said third-party applications that make purchases for customers should operate transparently and respect whether a business chooses to allow them on its platform. The company compared AI shopping agents to food delivery services working with restaurants or travel agencies booking flights through airlines.

“Agentic” apps such as Muse, Amazon argued, should follow the same principle. The company has therefore asked Meta to remove Amazon from Muse’s shopping experience.

Meta launched Muse on September 8 as a personal AI agent designed to handle a variety of everyday tasks, including sending emails, booking travel, shopping online, and helping users keep track of their goals. Since its release, the app has climbed to the number-one position on Apple’s U.S. App Store chart for free iPhone apps.

According to Meta, Muse operates through a secure virtual machine that contains the AI agent and the user’s data. The virtual machine includes its own browser, allowing Muse to visit websites and carry out tasks on a user’s behalf.

Amazon says Meta did not notify the company ahead of time that Muse would be accessing its website. The retailer also claims that Muse does not identify itself as an AI agent while shopping.

Amazon has raised concerns that this type of automated shopping could also bypass some of the personalization and other features that are part of its normal shopping experience.

Another concern involves customer credentials and account information. Because Muse does not identify itself to Amazon, the company argues that the AI effectively acts as an undisclosed third party accessing customer accounts without Amazon’s knowledge or permission.

Meta has pushed back on those concerns, saying Muse cannot directly see a user’s passwords or payment information. Instead, credentials provided by users are stored securely, allowing Muse to use them without exposing the actual information to the AI. Meta also says Muse requires user approval before carrying out actions such as completing a purchase.

The dispute is not Amazon’s first clash with an AI-powered shopping assistant.

In 2025, Amazon sued Perplexity over its Comet browser, seeking to prevent the AI-powered browser from shopping on Amazon. Amazon argued that Perplexity had violated the federal Computer Fraud and Abuse Act by accessing Amazon accounts without authorization.

A federal judge initially granted Amazon a preliminary injunction in March. However, in August, a three-judge panel of the Ninth Circuit Court of Appeals overturned that order and sent the case back to the district court.

The appeals court found that Amazon was unlikely to succeed in arguing that Perplexity itself was accessing Amazon’s computer systems. Instead, the judges concluded that the customer was the one accessing Amazon, while Comet was acting as a tool on the customer’s behalf.

Meta had not immediately responded to a request for comment at the time of publication.

The dispute highlights a growing challenge for online retailers as AI agents become capable of browsing websites and making purchases for users. While companies like Meta see these agents as a convenient way to automate everyday tasks, retailers such as Amazon are increasingly questioning who should control and authorize that automated access.