A federal appeals court is weighing how a decades-old U.S. hacking statute should apply to an AI tool accused of getting into Amazon customer accounts without permission, a dispute that could shape who bears responsibility when software acts on a user's behalf.
Perplexity AI's shopping-focused "agentic" system unlawfully accessed Amazon.com Inc.
(NASDAQ:AMZN) customer accounts without permission, potentially triggering liability under the 1986 Computer Fraud and Abuse Act, Reuters reported.
Judge John Hinderaker, sitting by designation on the panel, noted the statute was "not really built" with AI in mind and asked a central question running through the case: "Does an AI agent ever have intent?" Benzinga contacted Perplexity AI and Amazon for comment, but did not receive a response at the time of publication.
Read Also: 'Nightmare Of A Neighbor': Group Sues To Halt SpaceX Texas Refuge Land Deal The dispute stems from Perplexity's appeal of a lower court ruling that temporarily blocked it from deploying its AI agents for shopping tasks on Amazon.
