Amazon has filed a lawsuit against Perplexity AI, claiming the startup’s autonomous shopping tool accessed private customer accounts and made purchases disguised as human activity. The complaint, filed in the U.S. District Court for the Northern District of California, marks the latest clash between major tech players over the growing use of “agentic” AI systems in online commerce, according to Reuters.
In the filing, Amazon alleged that Perplexity’s Comet AI browser agent posed “serious security risks” by interacting with the Amazon Store as if it were a human user. The company said the tool bypassed technical safeguards and continued operating despite multiple cease-and-desist requests. “That Perplexity’s trespass involves code rather than a lockpick makes it no less unlawful,” the complaint stated.
Perplexity pushes back, calling Amazon’s move intimidation
Perplexity has denied any wrongdoing and described Amazon’s lawsuit as an attempt to suppress competition. The startup said that all user credentials remain stored locally on customer devices, not on its servers, and that Comet AI simply helps users compare products and automate purchases.
“Amazon doesn’t care about easier shopping — it cares about protecting its ad business,” Perplexity wrote in a blog post, accusing the retail giant of “bullying smaller innovators through legal threats.” The company also said Amazon’s stance threatens user choice and the future of AI assistants.
Broader debate on how far autonomous AI can go
The dispute underscores a larger question now facing the tech industry: how much freedom should AI agents have to navigate websites and act on behalf of users? While Amazon argues that third-party automation tools must respect security boundaries, Perplexity and other AI startups see autonomy as the next step in digital convenience.
Amazon itself is developing similar tools, including “Buy For Me”, which enables automated shopping across brands, and “Rufus”, an AI assistant designed to guide purchases within its ecosystem.
As both sides double down, the outcome of this lawsuit could set a precedent for how AI agents operate in commercial environments — and how far they can go before crossing legal and ethical lines.
