“They would only get in trouble if they summarized the story incorrectly and made it defamatory when it wasn’t before. “That’s something they would really be at legal risk for, especially if they don’t credit the original source clearly enough and people can’t easily go to that source to verify,” he says. “If Perplexity’s edits are what make the story defamatory, 230 doesn’t cover that, according to a large body of case law interpreting it.”
In one case observed by WIRED, the Perplexity chatbot falsely claimed, though prominently linking to the original source, that WIRED had reported that a specific police officer in California had committed a crime. (“We have been very candid that the answers will not be accurate 100% of the time and may cause hallucinations,” Srinivas said in response to questions for the story we published earlier this week, “but a central aspect “Our mission is to continue improving accuracy and user experience.”
“If you want to be formal,” Grimmelmann says, “I think this is a set of claims that would overcome a motion to dismiss based on a bunch of theories. I’m not saying he’ll win in the end, but if the facts bear out what Forbes and WIRED, the police officer (a group of potential plaintiffs) allege, are the kind of things that, if proven and other facts were bad for Perplexity, could give rise to responsibility.”
Not all experts agree with Grimmelmann. Pam Samuelson, a professor of law and information at UC Berkeley, writes in an email that copyright infringement is “the use of another person’s expression in a way that undermines the author’s ability to obtain adequate remuneration for the value of unauthorized use. A verbatim phrase is probably not an infringement.”
Bhamati Viswanathan, a member of the New England Law School, says she is skeptical that the summary passes the threshold of substantial similarity that is typically necessary for a successful infringement lawsuit, although she does not believe that is the end of the matter. “It certainly shouldn’t pass the smell test,” she wrote in an email. “I would say it should be enough to get your case over the motion to dismiss threshold, especially given all the signs you had that real things were being copied.”
However, he argues that focusing on the narrow technical merits of such claims may not be the right way to think about things, as technology companies can adjust their practices to respect the letter of antiquated copyright laws while still time flagrantly violate its purpose. She believes an entirely new legal framework may be necessary to correct market distortions and advance the underlying goals of U.S. copyright law, including allowing people to financially benefit from original creative work like journalism. so that they are encouraged to produce it. —with, in theory, benefits for society.
“In my opinion, there are strong arguments to support the intuition that generative AI relies on large-scale copyright infringement,” he writes. “The initial question is: where do we go from there? And the most important long-term question is: how do we ensure that creators and creative economies survive? Ironically, AI is teaching us that creativity is more valuable and in demand than ever. But even as we recognize this, we see the potential to undermine and ultimately gut the ecosystems that allow creators to make a living from their work. That is the riddle we must solve, not eventually, but now.”