Amazon.com Inc. has launched a federal lawsuit against up-and-coming AI firm Perplexity AI Inc., aiming to halt the startup’s innovative tool from automating purchases on its vast e-commerce platform.
The complaint, lodged in the U.S. District Court for the Northern District of California in San Francisco, charges Perplexity with computer fraud and breach of contract.
At the heart of the dispute is Comet, Perplexity’s advanced AI browser agent, which enables users to delegate online shopping tasks—including adding items to carts and completing transactions—without manual intervention.
Amazon contends that Comet operates covertly, masquerading as a standard human user via Google Chrome sessions, thereby circumventing site safeguards and flouting the retailer’s explicit terms of service.
The e-commerce behemoth issued a cease-and-desist demand to Perplexity last Friday, citing risks to user privacy and a diminished shopping interface for legitimate customers.
When the startup declined to comply, Amazon escalated to court, seeking an injunction to block Comet’s access entirely.
Sources close to the negotiations reveal that Amazon views the agent’s actions as akin to digital trespassing, emphasizing that “no entity, regardless of its technological sophistication, is permitted to ignore clear prohibitions.”
Perplexity, a darling of the AI investment scene with a $20 billion valuation, fired back forcefully.
A company representative dismissed the suit as evidence of Amazon’s monopolistic tendencies, while CEO Aravind Srinivas, in a pointed statement, lamented the “customer-unfriendly” stance.
In a recent company blog, Perplexity argued that empowering users with choice in AI assistants fosters innovation and convenience, accusing Amazon of wielding its market dominance to suppress rivals.
“This isn’t about security—it’s about control,” the post asserted, highlighting how third-party tools like Comet democratize access to seamless shopping.
The Rise of AI Agents: Convenience or Intrusion?
This high-stakes confrontation arrives amid a surge in “agentic” AI technologies—systems designed not just to answer queries or generate text, but to execute multi-step actions on behalf of users.
Perplexity’s Comet exemplifies this shift, integrating web navigation, decision-making, and transaction execution into a single, intuitive interface.
Users simply instruct the agent to “buy running shoes under $100 with good reviews,” and it handles the rest, from searching listings to finalizing payments via linked accounts.
Amazon, no stranger to AI experimentation, is simultaneously rolling out its own agent-based features.
Its “Buy For Me” tool, currently in beta, allows in-app purchases from external brand websites, while the Rufus assistant scours product catalogs, curates recommendations, and populates shopping carts.
Yet, the company insists on strict transparency for any external agents interacting with its ecosystem.
“Third-party apps must declare their automated nature and honor platform rules,” said Amazon spokesperson Lara Hendrickson. ”
Failure to do so erodes trust and complicates essential services like customer support.”
The timing is telling: In November 2024, Amazon successfully persuaded Perplexity to pause similar agent deployments following informal talks.
However, the reintroduction of Comet in August 2025—now equipped with evasion tactics against Amazon’s blocking efforts—prompted the renewed clash.
Perplexity maintains that its tool neither scrapes data for training nor mines content, confining activities to user-directed buys.
Srinivas, speaking to reporters, equated AI agents to empowered proxies: “If a human can shop on Amazon, so should the intelligent tools they trust. Policing the distinction stifles progress.”
Broader Ramifications for E-Commerce and AI Ethics
Legal experts predict the case, docketed as *Amazon.com Services LLC v. Perplexity AI Inc.* (3:25-cv-09514), could establish critical precedents.
Key questions include whether AI agents qualify as “users” under federal computer fraud statutes and how platforms can enforce bot detection without stifling legitimate automation.
The outcome may influence not only shopping but also sectors like travel booking and food delivery, where agentic tools promise efficiency gains.
For Amazon, the stakes extend to its core revenue streams. The company’s advertising empire, which generated billions last quarter through sponsored search placements, could face disruption if bots bypass traditional query funnels.
During a recent earnings discussion, CEO Andy Jassy acknowledged the nascent flaws in rival agents—such as impersonal recommendations and erratic logistics forecasts—but expressed openness to collaborations.
“We’re engaging with developers to align on standards that prioritize the shopper,” Jassy noted.
Complicating matters, Perplexity isn’t a pure adversary. The startup relies heavily on Amazon Web Services (AWS) for its infrastructure, with commitments reportedly exceeding hundreds of millions.
Jeff Bezos, Amazon’s founder, holds a personal stake in Perplexity, adding an ironic twist to the feud.
Despite past accolades at AWS events, where Perplexity was showcased as a cloud success story, the symbiosis appears strained.
Perplexity’s track record adds layers of scrutiny. The firm has weathered separate allegations from news publishers over unauthorized content use in AI summaries and from Reddit regarding illicit data harvesting.
Undeterred, it champions open access to information, vowing to defend user autonomy in court.
As the litigation unfolds, industry watchers anticipate ripple effects. “This isn’t isolated—it’s a bellwether for how Big Tech polices the AI frontier,” said one analyst.
For consumers, the battle pits the allure of hands-free shopping against concerns over data security and platform integrity.
