Amazon bars Meta's Muse AI from shopping on its site

Amazon has prevented Meta's newly released AI agent from making purchases on its platform, reportedly over privacy and security concerns.

Amazon blocks unauthorized AI purchases. Shoppers who try to use Meta's Muse on Amazon now encounter a notice declaring that continued access by an unauthorized AI agent violates Amazon's Conditions of Use. The retail giant first requested that Meta voluntarily keep its site out of the agent's reach.

An Amazon spokesperson said that third-party tools offering to buy goods for customers from rival businesses ought to act transparently and honor each provider's choice on whether to allow such activity.

Meta launched Muse earlier this month. The social-media company presented Muse as a "secure, private personal AI agent that proactively helps with people's goals and suggests ideas." It is capable of completing forms, dispatching emails, and handling transactions through a one-time virtual card supplied by Link by Stripe.

Amazon cites privacy and security concerns. The retailer is reportedly frustrated that Muse does not identify itself while navigating the website. Amazon also maintains that the agent creates privacy and security risks by gaining access to and storing customer data.

Meta disputes those assertions. In its launch announcement, the company stated, "Muse has no visibility into people's passwords or payment methods. Any credentials a person shares go into secure storage, so Muse can use them without seeing them, including passwords a person types into the browser themselves."

AI companies compete over online turf. The clash illustrates mounting friction among technology giants as AI assistants expand their interactions with external services. Engadget has reached out to Amazon and Meta for comment on the matter.