What Actually Happened

CNN filed suit against Perplexity AI on May 28, 2026, in the U.S. District Court for the Southern District of New York. According to reporting by Variety, Engadget, and CNN's own Brian Stelter, the 54-page complaint accuses Perplexity of scraping more than 17,000 CNN stories, photos, videos, and other content — then feeding that material into its AI products without a license, without payment, and without permission.

The lawsuit also alleges Perplexity reproduced verbatim copies of paywalled CNN articles in its query responses. Paying subscribers only. Content CNN charges money for. Perplexity was apparently just handing it out for free.

CNN says this is the network's first legal action against an AI company, and according to Variety, it is believed to be the first copyright lawsuit by any television network against an AI firm.

The Failed Deal Nobody's Talking About Enough

CNN and Perplexity were in active negotiations last year for a licensing deal. The arrangement would have given Perplexity's paid subscribers access to paywalled CNN content — a straightforward commercial arrangement.

The deal fell apart. That happens in business.

What's notable: continuing to use the other party's content after negotiations collapse and legal warnings arrive. According to CNN's lawsuit, that is exactly what Perplexity did. CNN's legal team sent warnings. Perplexity never responded.

Perplexity knew CNN had not granted permission. CNN explicitly said so. Perplexity kept scraping anyway.

Perplexity's Defense Is Weaker Than It Sounds

Perplexity Chief Communications Officer Jesse Dwyer gave the same one-line response to CNN, NPR, and Variety: "You can't copyright facts."

Copyright law doesn't protect just facts — it protects the original expression of those facts. The specific words a journalist writes. The structure of a story. The editorial choices made in crafting an article. CNN isn't claiming Perplexity stole the fact that a hurricane happened. It's claiming Perplexity copied the actual articles, word for word, including paywalled ones.

Dwyer's response is a talking point. Real lawyers will need much stronger arguments in court.

The lawsuit also accuses Perplexity of trademark violations — specifically, that Perplexity's AI tools attributed fabricated, "hallucinated" content to CNN. Perplexity copied real CNN content without permission and also invented fake CNN content and attached CNN's name to it.

The Bigger Picture

CNN isn't alone. According to NPR and Engadget, the following have also sued Perplexity: The New York Times, The New York Post, Dow Jones (publisher of the Wall Street Journal), the Chicago Tribune, Reddit, Merriam-Webster, Encyclopedia Britannica, and Nikkei.

That is a significant pile of litigation for a company that, according to Engadget, is valued at tens of billions of dollars.

Meanwhile, some publishers took a different approach. According to Variety, Time and USA Today's parent company, Gannett, struck licensing deals with Perplexity instead of suing. Meta signed a separate content-licensing deal with CNN in December 2024 — proof that these arrangements are possible when companies negotiate in good faith.

The industry is splitting into two camps: those who sue and those who deal. Both are legitimate strategies. What is NOT legitimate is a third option Perplexity apparently preferred — take the content, ignore the lawyers, and dare someone to stop you.

What Mainstream Coverage Is Getting Wrong

Most outlets are framing this as a standard tech-vs.-media copyright fight. The failed licensing negotiation detail is critical and keeps getting minimized. This wasn't a case of Perplexity being unaware of CNN's position. Perplexity knew. It knew CNN wanted a deal. It knew CNN's lawyers objected. It kept going.

Also underreported: the hallucination angle. Perplexity putting fabricated content under CNN's name isn't just a copyright issue — it's a brand integrity issue and potentially a defamation issue depending on what was fabricated.

What This Means for Regular People

If you use Perplexity — or any AI search tool — to get news summaries, understand what you're getting: content that may have been taken without permission from outlets that pay reporters, editors, and photographers to produce it.

Quality journalism is expensive. War correspondents, investigative reporters, data teams — none of that is free. When AI companies vacuum up that work and redistribute it without paying, they avoid licensing costs that original reporting requires.

You may not love CNN. But the principle applies to every news organization — left, right, or center. If AI companies can strip-mine content without consequence, the reporters who produce it don't get paid. Then the content stops existing.

Perplexity is valued at tens of billions of dollars. It can afford to license content. It chose not to. Now it's going to court to explain why.