Amazon blocks Meta's Muse AI agent from accessing its store

Amazon has taken steps to prevent Meta's Muse AI shopping assistant from accessing its platform and making purchases on behalf of users. Users of the service began encountering a notification on Sunday explaining that the AI agent's access breaches Amazon's terms of service.

The Block and Amazon's Stance#

The e-commerce giant raised two main objections to Muse's operation. First, Meta apparently did not seek Amazon's approval before allowing Muse to shop on its store. Second, Amazon cited concerns about the agent's ability to capture customer login credentials and its failure to properly identify itself while browsing, creating potential security and privacy gaps.

"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," Amazon stated.

Meta's Claims vs. Emerging Concerns#

When Meta launched Muse earlier this month, the company asserted that the agent cannot access sensitive login credentials or payment card details. However, privacy issues have surfaced since its release, with security researchers discovering that Muse can read the contents of user messages even without the appropriate permission settings being enabled — a finding that undercuts Meta's security guarantees.

Part of Amazon's Broader Defensive Strategy#

This move reflects Amazon's ongoing campaign to limit third-party AI services from accessing its ecosystem. In November, Amazon sued Perplexity, the AI search provider, over its Comet shopping tool, though a federal judge ultimately sided with Perplexity in August 2026. Additionally, Amazon has stripped down order confirmation emails starting in July, removing product images and specific item names to prevent external AI systems from harvesting that data for competitive shopping experiences.