The landscape for artificial intelligence applications in e-commerce shifted significantly on Tuesday when the San Francisco-based 9th U.S. Circuit Court of Appeals struck down a temporary ban that had prohibited Perplexity from deploying its AI-powered shopping agents on Amazon's platform. This ruling marks a watershed moment in how courts view the intersection of autonomous AI systems and retail platforms, establishing the first federal appeals court precedent for whether AI agents acting as user proxies can legally navigate commercial websites.
Amazon had initiated legal action against Perplexity in November, asserting that the company's Comet browser and associated AI agent were making unauthorized access to private customer accounts. According to Amazon's complaint, Perplexity's technology could log into user shopping accounts and execute purchases autonomously, creating security vulnerabilities and flouting Amazon's repeated requests for the company to cease operations. Amazon contended the arrangement violated the Computer Fraud and Abuse Act, a foundational statute governing unauthorized computer access in the United States.
Perplexity's response challenged the fundamental framing of the lawsuit. The AI startup argued that the legal action lacked substantive merit and represented a strategic effort to prevent Amazon users from leveraging Comet because autonomous agents "don't have eyeballs to see the pervasive advertising Amazon bombards its users with." This counterargument introduced a competitive dimension to the dispute, suggesting Amazon's objections were less about security and more about protecting its advertising revenue model and user engagement metrics.
In March, a federal judge in California sided with Amazon's preliminary position, issuing a temporary injunction against Perplexity. The judge determined that Amazon had demonstrated "strong evidence" that Perplexity's AI agent violated the Computer Fraud and Abuse Act, warranting the restraining order. This decision appeared to establish that the startup faced an uphill battle and suggested courts would view autonomous access to shopping platforms as legally problematic.
The appeals court's reversal on Tuesday fundamentally reframed this analysis. The three-judge panel determined that Amazon was unlikely to ultimately succeed in its claim, based on a crucial technical distinction embedded in the Computer Fraud and Abuse Act. The statute prohibits unauthorized access to computers, but the appeals court reasoned that Perplexity itself was not the party accessing Amazon's systems. Rather, it was the individual users operating the AI agents who constituted the "accessors" under the law's language. Because these users possessed authorization to access their own accounts, the court found no statutory violation had occurred.
This interpretation carries substantial implications for the emerging category of agentic AI systems—sophisticated tools capable of planning, reasoning, and executing complex tasks with minimal direct human supervision. Unlike traditional chatbots that respond to individual queries, agentic AI can autonomously navigate websites, make decisions, and conduct transactions across multiple platforms. The ruling creates legal breathing room for this technology to proliferate across commercial platforms without facing the kind of immediate restraint Amazon's lawsuit sought to impose.
Amazon's response indicated the company would continue contesting the matter. A company spokesperson stated that Amazon remained "confident in our case and are evaluating our next steps," signaling potential further appeals or adjusted legal strategies. This determination suggests the dispute may not conclude with the appeals court's ruling, particularly if Amazon believes the decision misinterprets congressional intent regarding unauthorized platform access.
Perplexity's statement characterized the decision as vindicating user rights. Company spokesperson Jesse Dwyer asserted that Perplexity would "continue to fight for the right of internet users to choose whatever AI they want," framing the case as a broader struggle against limitations on consumer choice and technological autonomy. This rhetoric positions the startup as defending user interests against what it portrays as corporate gatekeeping.
For Malaysian and Southeast Asian technology observers, this ruling warrants close attention because it establishes precedent that will likely influence how AI tools are regulated and deployed throughout the region. As agentic AI systems gain sophistication and adoption accelerates, regulators and courts across Asia will face similar questions about whether autonomous agents should enjoy the same platform access rights as their human operators. The decision also highlights tensions between technology innovation and corporate interests in controlling user data and advertising exposure.
The case ultimately pivots on interpretive questions that extend beyond Perplexity and Amazon. Courts globally must determine whether authorization inheres to users or to the tools they deploy, whether corporate terms of service can override statutory frameworks, and how regulatory regimes should accommodate user choice versus platform security. The appeals court's emphasis on user authorization rather than corporate intent suggests courts may prove receptive to agentic AI applications where genuine users remain in control, even when platforms object.
Looking forward, this decision creates a framework within which agentic AI developers can argue for greater access to major platforms based on user authorization principles. However, the ruling does not foreclose other legal theories Amazon or other companies might pursue, including breach of contract claims, intellectual property allegations, or statutes specifically addressing scraping and data extraction. The appellate decision may thus represent only an early skirmish in a protracted legal conflict over how autonomous AI systems interact with controlled digital environments.
