CNN filed a copyright and trademark lawsuit against Perplexity in the United States District Court for the Southern District of New York, alleging the AI company "unlawfully copying and distributing" CNN's reporting and using its marks without permission, according to CNN and reporting in Variety. The complaint, as reported by Variety, alleges Perplexity scraped more than 17,000 CNN stories, photos and videos and used them to train its products. Perplexity responded to media requests with the quoted line "You can't copyright facts," per TheWrap. CNN says it attempted licensing talks with Perplexity last year but did not reach terms, and the suit joins multiple publisher actions against Perplexity even as other outlets have negotiated licensing deals, reporting by CNN and other outlets states.
What happened
CNN filed a federal copyright and trademark suit against Perplexity in the United States District Court for the Southern District of New York, alleging the AI company unlawfully copied and distributed CNN content without permission, according to CNN's reporting and contemporaneous coverage by Variety and TheWrap.
Allegations in the complaint
Per the 54-page complaint described by Variety, CNN alleges Perplexity scraped more than 17,000 of the network's stories, photos, videos and other content and used that material in training or operating its services. CNN's public statement included the quote that Perplexity "should not be able to steal from entities that create the original content Perplexity exploits," as reported by CNN and echoed in aggregator coverage.
Perplexity's public response
In response to media inquiries, Perplexity's comment to TheWrap was published as "You can't copyright facts." Variety also reports the complaint points to Perplexity advertising that it helps users skip the "extra steps and clicks" normally needed to access full news articles.
Editorial analysis - technical context
Industry-pattern observations show the current legal debate centers on the boundary between uncopyrightable facts and copyrighted expressive content, and on whether large language models or retrieval layers reproduce protectable expression. For practitioners, that distinction affects dataset curation choices: whether to rely on licensed publisher feeds, focus on metadata and links, or implement redaction and copyright-safe filtering when ingesting news content.
Industry context
Publishers have pursued a two-track response to AI news aggregation, combining litigation with negotiated licensing. CNN's suit follows actions by other publishers and sits alongside licensing deals some outlets have reached with AI firms, as documented in CNN and Variety coverage. That split is reshaping commercial terms for content access and is already prompting public scrutiny of how chatbots present source attributions.
What to watch
Observers should track court filings and discovery for specifics about dataset sources and ingestion methods, licensing outcomes between publishers and AI firms, and any early judicial guidance on whether training or verbatim reproduction of news content violates copyright. For practitioners: follow posted pleadings for technical exhibits and deposition topics that could clarify acceptable ingestion and summarization practices in product design.
Reported-scope caveat
The factual claims above come from the published complaint and reporting by CNN, Variety and TheWrap; Perplexity's full legal answer or discovery has not been made public as of these reports.
Why it matters
The case brings a television-network plaintiff into the growing set of copyright disputes over generative-AI sourcing, and could influence commercial licensing, dataset hygiene, and product UX for systems that summarize or republish news content.
Scoring Rationale #
This is a notable legal action that could shape precedent around training data and content licensing for AI systems; it matters to engineers and product managers who build news-aware models and ingestion pipelines. The case is part of a broader wave of publisher litigation and licensing activity.
Practice interview problems based on real data
1,500+ SQL & Python problems across 15 industry datasets — the exact type of data you work with.