CNN filed a copyright and trademark lawsuit against Perplexity AI on May 28, 2026 in the US District Court for the Southern District of New York, alleging the AI search company unlawfully scraped and redistributed more than 17,000 of its news stories, photos, and videos.

The case (Cable News Network Inc v. Perplexity AI, Inc., 1:26-cv-04427) is CNN’s first AI copyright action and is the first such lawsuit filed by any television network. It marks a significant escalation in the broader conflict between established media companies and AI platforms over the right to train and operate on copyrighted content.

What happened

CNN says it sought a licensing agreement with Perplexity last year but the two companies could not agree on terms, including restrictions on how Perplexity’s chatbot could use CNN’s journalism. After negotiations broke down, CNN blocked Perplexity’s scraping bot from accessing its content via robots.txt settings.

Perplexity continued to use CNN’s content regardless. The lawsuit alleges the company’s AI search product reproduces, paraphrases, and distributes CNN’s original reporting in ways that substitute for visiting CNN’s own site, reducing traffic and undermining CNN’s advertising-driven business model.

Perplexity’s chief communications officer Jesse Dwyer responded with a blunt statement: “You can’t copyright facts.”

CNN is seeking statutory damages and a court order stopping Perplexity from using its content.

Nine publishers now in active litigation

CNN is not alone. As of May 31, 2026, nine organisations have filed active suits against Perplexity: The New York Times, News Corp and Dow Jones, the New York Post, the Chicago Tribune, Encyclopedia Britannica, Merriam-Webster, Reddit, and Japan’s Yomiuri Shimbun.

The scale of this legal front is significant. Every major category of content creator is now represented: newspapers, broadcast media, reference publishers, and user-generated platforms. The message is consistent: AI companies cannot build commercial products on top of content they did not create and did not pay for.

What This Means for Business

There are three practical implications for businesses watching this space.

The legal landscape around AI tools is hardening. For a long time, AI companies operated in a grey zone where the legality of training on web content was genuinely unclear. That grey zone is narrowing fast. Courts are being asked to define what AI companies can and cannot do with published content, and the answers will shape what products can be built and offered commercially. If you are using AI tools that aggregate and synthesise news or third-party content, the terms of those tools may need to change as this litigation plays out.

Content creators have real and immediate exposure. If your business produces original content, that content may already be scraped and used by AI systems without your knowledge or consent. Checking your robots.txt settings, monitoring how AI tools cite or paraphrase your material, and reviewing whether your vendor agreements address AI usage are all steps worth taking now rather than after a problem surfaces.

AI licensing is becoming a genuine cost line. Several AI companies, including OpenAI and Google, have signed content licensing deals with major publishers. If courts rule against Perplexity’s position and require licensing for training and inference, the cost of operating AI products will rise. Those costs will eventually reach enterprise customers. The era of AI tools built on freely scraped content may be shorter than the optimists assumed.

For businesses evaluating AI search and research tools, vendor legal stability matters more than it used to. A product facing nine simultaneous infringement suits is carrying legal risk that could change what it can offer or at what cost.

Perplexity’s “you can’t copyright facts” defence has some narrow merit when applied to bare factual statements. It has far less when applied to original video footage, photography, and the editorial judgment that shapes how facts are selected and presented. CNN’s case will likely test exactly where that line falls.

The outcome will matter well beyond media and AI. It will help define what businesses can and cannot do with AI tools that pull from the open web, and how much the underlying data infrastructure that powers those tools actually costs.