CNN filed a federal lawsuit against AI startup Perplexity on May 28, 2026, accusing the company of "massive copyright infringement." The network alleges that Perplexity scraped more than 17,000 CNN stories, photos, and videos without authorization — and used that material to train its AI products.
What Happened
The 54-page complaint was filed in the U.S. District Court for the Southern District of New York. CNN claims that Perplexity systematically bypassed site protections to collect content without a licensing agreement. The lawsuit also includes a trademark infringement claim: Perplexity allegedly advertised a nonexistent "partnership" with CNN through its Comet Plus subscription tier.
Licensing talks between the two companies previously broke down. CNN then blocked Perplexity's scraping bot. According to the complaint, the startup continued collecting data by working around those restrictions.
Why It Matters
CNN is the first television network to sue an AI company over copyright infringement. Until now, similar lawsuits came solely from print publishers — the New York Times, Dow Jones, and the New York Post. If the court rules in CNN's favor, the precedent will affect the entire AI search market and any startup built on synthesizing third-party content.
Both Sides and Broader Context
Perplexity's response was brief: a company spokesperson said, "You can't copyright facts." That's a standard industry argument, but legal experts note that CNN's claims go beyond facts — they cover original reporting, editorial choices, and media files, which are exactly what copyright law protects.
- NYT vs. OpenAI — similar case in court since 2023
- Dow Jones and NY Post vs. Perplexity — already pending
- Time and USA Today — chose to negotiate licensing deals instead
What Comes Next
CNN is seeking statutory damages and a permanent injunction barring Perplexity from using its content. The outcome could set the standard for the entire industry. A publisher win would force AI companies to negotiate licensing deals at scale — and fundamentally reshape the economics of generative AI. A Perplexity win would leave the open web as fair game for training data with few restrictions.
