CNN sues Perplexity over alleged copyright infringement

TL;DR:

  • CNN has filed suit against AI search firm Perplexity in New York federal court, alleging it unlawfully copied thousands of stories, videos and images and distributes “identical or substantially similar” content.
  • Perplexity responded that “you can’t copyright facts”; CNN, owned by Warner Bros. Discovery, is seeking unspecified damages and an injunction.
  • The case adds to dozens of US suits by publishers and authors against AI firms, and Perplexity already faces actions from the New York Times, Reddit and Dow Jones.

The dispute over who owns the content feeding AI answers has gained another front-page plaintiff. CNN alleges Perplexity, which uses AI to scour the web and answer queries, copied its journalism wholesale to power a competing product that undermines “the economic incentives that make original newsgathering possible”. Perplexity’s defence — that facts are not copyrightable — previews the central legal question these cases will turn on.

Part of a broader reckoning

CNN’s suit is one of dozens of high-stakes US cases brought since ChatGPT’s 2022 launch, as publishers, authors and rights holders test whether training and answering on copyrighted work constitutes infringement. The outcomes are diverging. Some firms have litigated; others have signed licensing deals that compensate publishers and link back to original articles. Anthropic became the first AI company to settle one of these cases last year, agreeing to pay $1.5bn to resolve a class action from authors — a figure that now anchors expectations for what exposure looks like.

Perplexity in particular has become a magnet for litigation, with the New York Times, Reddit and Dow Jones among those already suing. The clustering of cases against a single firm suggests publishers see its answer-engine model — which can satisfy a query without sending traffic onward — as more directly threatening than a chatbot that merely trained on their archives.

Looking forward

For UK media and AI businesses, the US litigation is a preview rather than a sideshow. British publishers face the same tension between licensing their archives and seeing them ingested for free, and UK courts and the government’s evolving text-and-data-mining stance will eventually have to resolve similar questions. Whether the dominant outcome becomes licensing markets or court-imposed limits will shape how AI search products can legally operate on both sides of the Atlantic.

Share this article