Meta’s Muse joins ChatGPT and Claude in Amazon shopping ban

Amazon is having none of third-party AI assistants shopping on its website or even browsing it on users' behalf. Meta’s Muse ban is just the most recent addition to a long list.

Meta introduced Muse as the “first personal AI agent” that can browse, fill out forms, negotiate on the user's behalf, and make purchases, with free coverage for damage, loss, price drops, and other issues up to $1,000 per claim.

But not on Amazon. GeekWire reported that Amazon began blocking Muse’s access to its website. Users report seeing an error message.

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

Amazon confirmed it requested that Meta remove Amazon from the Muse experience, and explained that Meta never asked for permission for Muse to access its store.

"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. This helps ensure a safe, secure, and reliable customer experience,” the tech giant explains in a statement.

Most AI assistants are banned on Amazon, but not its own

Muse is not the first to fall in an arms race with Amazon.

Other chatbots, like ChatGPT and Claude, confirm they’re unable to even check prices on Amazon directly and explain they’re being blocked from direct automated access.

Amazon has clearly laid out its policy in the robots.txt file, which tells crawlers and bots which pages they can and cannot access.

This file explicitly blocks user agents by name. The long list includes signatures tied to Claude, ChatGPT, Copilot, Perplexity, Manus, Grok, and Google’s AI crawlers – each getting a blank “Disallow.” This means that AI assistants are shut out of the entire Amazon website.

Separately, any compliant bot is disallowed from accessing transactional endpoints, such as sign-in, cart, wishlist, rate-this-item, and others.

Robots.txt relies on good-faith compliance, and rogue bots might still disrespect it. Services often use an arsenal of other server-side defenses, such as a web application firewall (WAF), rate limiting, CAPTCHA checks, and others.

However, Amazon itself offers users its own AI models to help with purchases, including in other stores.

Previously, the e-commerce giant introduced the Rufus chatbot as a shopping assistant built directly into the app and website. Later, Amazon axed it and unified AI assistance under “Alexa for Shopping,” a bot that can answer queries and take actions on users' behalf.

Not only can Alexa “effortlessly add items” to the cart, but it can also search for products on other shops using an agentic AI feature called Buy for Me.

“Buy for Me agentic AI feature handles the entire purchase on your behalf using your primary address and credit card,” Amazon says.

Amazon has previously been taking action against other AI assistants. For example, it’s currently locked in a legal battle with Perplexity over its AI shopping features in the Comet browser. Amazon sued Perplexity, accusing the startup of covertly accessing private Amazon customer accounts, Reuters reported.

The US appeals court overturned a ruling that initially barred Perplexity from using AI-powered tools on Amazon, but Amazon remained confident in its case, saying it “respectfully disagrees” with the decision.

Meta’s Muse continues to work on other stores. After a successful model launch, Meta stock soared 11% on Monday.