CNN joins growing legal fight against Perplexity

CNN escalates the media industry’s conflict with AI search

CNN has sued Perplexity, adding another major publisher to the widening legal campaign against AI companies accused of scraping and reproducing journalism without permission. According to the supplied source text, the network alleges “massive copyright infringement,” including the wrongful scraping, copying, and distribution of more than 17,000 pieces of CNN content.

The case matters because it sharpens several disputes at once: whether AI search products can ingest and summarize news content at scale without licensing it, whether reproducing article text in responses crosses clear copyright lines, and whether errors generated by those systems can create additional trademark or reputational harm.

The core allegations

CNN’s complaint says Perplexity unlawfully crawls, scrapes, copies, and distributes content from CNN’s digital platforms and third-party platforms. The lawsuit also alleges that Perplexity reproduced verbatim copies of CNN articles, including paywalled stories, inside answers returned to users.

That is an important distinction. Many AI copyright disputes revolve around training data, where the contested use happens behind the scenes. CNN’s allegations go further into product behavior by asserting that the service surfaced article content directly in user-facing responses. If proven, that would make the dispute less about distant model development practices and more about substitution in the market for current news.

The lawsuit also claims Perplexity attributed hallucinated material to CNN, which the network says violates its trademark. That adds a second dimension to the case. It is not only about uncompensated use of original reporting, but also about the risks that arise when an AI system places a trusted news brand next to information the publisher did not produce.

Part of a broader litigation wave

CNN is not acting alone. The source text notes that The New York Times, Chicago Tribune, Reddit, Merriam-Webster, Encyclopedia Britannica, and Nikkei have also filed lawsuits against Perplexity. That growing list shows that the conflict is expanding across different types of rightsholders, from newspapers to reference publishers to online communities.

Taken together, the cases suggest publishers increasingly see litigation as one of the few available tools to force boundaries around AI data use and answer generation. For media organizations, the economic concern is obvious. Original journalism is expensive to produce, but AI products may be able to capture user attention and advertising value by repackaging that work in a different interface.

The brand concern is just as significant. If a response is wrong but presented as drawing from a publisher’s reporting, the publisher can absorb reputational damage without controlling the output.

Failed licensing talks raise the stakes

The source material says CNN and Perplexity had at one point discussed a deal that would have allowed some paywalled CNN content to be made available to Perplexity’s paid subscribers. The agreement did not happen. CNN alleges that despite warnings from its legal team, Perplexity continued using CNN’s name and content in its products.

That detail matters because it reframes the dispute as more than a simple misunderstanding about web crawling norms. It suggests the companies had already recognized the commercial value of a licensed arrangement and still failed to reach one. In legal and business terms, that can harden positions quickly.

Perplexity’s public response in the source text is concise: “You can’t copyright facts.” That line captures a real principle, but it does not settle the issues in dispute. Copyright law may not protect bare facts, yet article structure, phrasing, selection, arrangement, and verbatim copying can still matter greatly, especially when the output allegedly includes near-direct reproductions of paywalled text.

Why this case is important

The lawsuit lands at a time when AI search products are trying to define themselves as alternatives to conventional web search and as more efficient ways to access information. Their value proposition depends on answering user questions directly rather than simply linking out. That same design creates tension with publishers whose businesses still rely on traffic, subscriptions, and control over how their reporting appears.

As more cases accumulate, courts may be pushed to draw clearer lines around what counts as transformative AI-assisted reference and what counts as market-harming reproduction. CNN’s complaint gives that debate a high-profile new test case, especially because it focuses on both copyright and trademark-style harms.

The immediate outcome is uncertain, but the direction is clear. Licensing, litigation, and product redesign are becoming central features of the relationship between AI companies and the news industry. CNN’s filing reinforces that publishers are no longer treating unauthorized AI use of journalism as an edge issue. They are increasingly treating it as a core business and legal threat.

This article is based on reporting by Engadget. Read the original article.