Amazon Blocks Meta's Muse AI Agent Over Unauthorized Automated E-Commerce Access
Amazon has blocked Meta's new autonomous AI agent, Muse, from making purchases on its platform on behalf of users without explicit authorization. The blockage follows unsuccessful negotiations urging Meta to exclude Amazon's site, triggering popup warnings citing terms-of-service violations for unauthorized AI access. The dispute underscores a growing conflict in agentic e-commerce over who controls the consumer shopping experience, privacy, and merchant-customer relationships. It also poses a threat to platform business models like Amazon's, which heavily rely on direct user browsing and sponsored advertising revenues. Meta's Muse operates inside an isolated virtual machine and navigates sites like a human browser when no public API exists, using stored user credentials without declaring its agent identity. Amazon argues that this undisclosed automated access poses privacy risks, violates site agreements, and deprives merchants of explicit consent.
## BACKGROUND
Agentic commerce refers to an emerging e-commerce paradigm where autonomous AI agents research, select, and purchase goods on behalf of consumers with minimal real-time human input. Unlike standard API integrations, agents like Muse simulate human browsing actions to interact with web services that lack open developer interfaces. Amazon has previously blocked similar automated shopping agents and web scrapers from AI providers like OpenAI, Google, and Perplexity to safeguard its proprietary platform ecosystem.