CNN sues Perplexity, accusing AI startup of copying thousands of stories verbatim
CNN alleges the AI startup scraped more than 17,000 articles and videos, reproduced some content verbatim, and bypassed the network's subscription paywall after licensing talks failed.
What matters
- CNN sued Perplexity in New York for allegedly scraping over 17,000 stories, photos, and videos.
- The suit claims Perplexity reproduces CNN content verbatim and bypasses subscription paywalls.
- It is believed to be the first AI copyright lawsuit filed by a television network.
- CNN attempted to negotiate a licensing deal with Perplexity before filing suit.
- Other publishers are split between suing (NYT, Dow Jones) and licensing (Time, Gannett, Meta).
What happened
On Thursday, CNN filed a lawsuit in the U.S. District Court for the Southern District of New York against Perplexity, the AI search startup reportedly valued in the tens of billions. The complaint accuses Perplexity of scraping more than 17,000 CNN stories, photographs, videos, and other media to train its AI "answer" engine. According to the network, Perplexity's tools not only ingest this material but also generate "verbatim" copies of CNN's work and provide users with information that is normally locked behind a CNN subscription.
The suit marks CNN's first legal action against an AI company over copyright infringement. It is also believed to be the first such litigation filed by a television network against a generative AI firm. CNN said it attempted to negotiate a licensing agreement with Perplexity before filing suit, but the companies could not reach terms. In a statement, a CNN spokesperson said the network "actively embraces the opportunities AI creates," but insisted that Perplexity "should not be able to steal from entities that create the original content" the company "exploits."
The complaint frames Perplexity's behavior as both copyright infringement and trademark violation, alleging that the startup distributes the network's content without consent. Perplexity, which markets itself as an AI-powered alternative to traditional search engines, has positioned citation and source transparency as core features. Yet CNN contends that citing a source does not grant permission to reproduce entire articles or circumvent payment systems.
Why it matters
CNN's complaint joins a crowded docket of publisher lawsuits targeting AI startups. The New York Times, Dow Jones (publisher of The Wall Street Journal), and the New York Post have already sued Perplexity over similar allegations of unauthorized content use. At the same time, a separate cohort of publishers—including Time, USA Today parent Gannett, and Meta—have pursued licensing deals, with Meta announcing a content-licensing agreement with CNN last December.
The divergence reflects an industry wrestling with a fundamental shift: as chatbots and AI search engines become primary gateways for news, publishers are fighting to ensure they are compensated. The lawsuit is particularly notable because television networks produce a mix of text, video, and imagery, raising distinct copyright questions that differ from text-only newspaper cases. If CNN can prove that Perplexity reproduced its content verbatim or bypassed paywalls, it could strengthen the argument that AI summarization and training go beyond fair use, potentially forcing more AI firms to the licensing table.
The case arrives as courts have yet to establish clear precedent on whether training generative AI models on copyrighted news material constitutes fair use. For Perplexity, a ruling against it could mean substantial damages and a forced restructuring of how its engine retrieves and synthesizes information. For publishers, a victory would validate the aggressive litigation strategy and likely increase the price of future licensing agreements across the industry.
Public reaction
No strong public signal was available from Reddit or broader social discussion at the time of publication.
What to watch
Legal observers will be watching whether Perplexity challenges the "verbatim" copying claim or argues that its outputs constitute transformative fair use. The case may also encourage other broadcast and cable networks to file similar suits, expanding the battlefield beyond print newspapers. Additionally, the paywall-bypass allegation could attract regulatory scrutiny beyond copyright law. If the court sides with CNN, the ruling could accelerate the fragmentation of the AI content ecosystem into licensed partners and litigious holdouts.
Sources
Public reaction
No significant Reddit or public discussion signals were captured in the available inputs.
Signals
- No strong public signal available
Open questions
- Will other TV networks follow CNN's lead?
- Can Perplexity prove its use qualifies as fair use?
- Will the lawsuit force Perplexity to change its content sourcing model?
What to do next
Developers
Audit your AI training data pipelines to ensure you have clear rights to use publisher content, and implement robust attribution and paywall-respecting scraping protocols.
Verbatim reproduction and paywall bypass are central to CNN's claims; technical architecture must minimize direct liability.
Founders
Treat content licensing as a core business risk; begin publisher negotiations early to avoid litigation that can stall fundraising or partnerships.
CNN's failed licensing talks preceded the lawsuit, suggesting that unresolved rights issues become existential threats as companies scale.
PMs
Design AI features that summarize without reproducing verbatim passages, and evaluate paywall bypass risks in your retrieval architecture.
Product decisions around output length and source access directly affect copyright exposure.
Investors
Factor potential copyright liabilities and licensing costs into valuation models for AI startups that rely on scraped news and media content.
A wave of publisher lawsuits could impose sudden cash costs or force business model changes on portfolio companies.
Operators
Review your organization's use of AI search tools for internal news consumption to ensure compliance with subscription terms and avoid indirect liability.
Using tools that bypass paywalls can expose companies to terms-of-service violations and reputational risk.
Testing notes
Caveats
- This story concerns ongoing litigation and business practices, not a released product, API, or model. There is no consumer-facing feature to test.