Answer-First Lead

CNN filed a 54-page copyright and trademark lawsuit against Perplexity AI in the Southern District of New York on May 28, alleging the AI search company scraped more than 17,000 articles, photos, and videos from CNN’s platforms and redistributed them — including verbatim reproduction of paywalled content. CNN is the ninth major publisher to sue Perplexity, and the first television network to take an AI company to court over copyright. Perplexity’s response: “You can’t copyright facts.”

🔍 THE BOTTOM LINE

This lawsuit isn’t really about 17,000 articles. It’s about whether AI companies have to pay for the raw material they consume — and whether “fair use” still means anything when the user never visits the original source.

What CNN Is Alleging

The complaint makes three core claims:

1. Massive unauthorized scraping. CNN alleges Perplexity “unlawfully crawls, scrapes, copies, and distributes CNN’s content from CNN Digital Platforms and third-party platforms.” CNN says it tried to block Perplexity’s unidentified crawlers — they got through anyway.

2. Verbatim reproduction. According to the complaint, prompting Perplexity’s AI search tool with the title of a CNN article produced “substantial verbatim portions” of the original piece. Not summaries. Not paraphrases. Near-copies. Including content behind CNN’s subscription paywall — meaning Perplexity was allegedly serving up paywalled journalism for free.

3. Trademark violation through hallucination. CNN claims Perplexity’s tools have incorrectly attributed “hallucinated” content to CNN — fabricating stories and slapping the CNN brand on them. That’s a trademark claim layered on top of copyright.

What is AI content scraping? AI content scraping is the automated extraction of text, images, and other media from websites to train AI models or power AI-generated responses. It works by using web crawlers — sometimes ignoring robots.txt and other opt-out signals — to harvest content at scale. For example, Perplexity’s crawlers allegedly continued scraping CNN content even after CNN attempted to block them.

The Failed Deal That Made It Worse

Here’s the detail that makes this messier: CNN and Perplexity were actually in negotiations last year. The proposed deal would have made paywalled CNN content available to Perplexity’s paid subscribers through the Comet browser, launched in October 2025.

The deal fell apart in November 2025. According to the lawsuit, the sticking point was “limits on Perplexity’s use of CNN content in its answers to users” — in other words, CNN wanted boundaries on how its content could be repackaged, and Perplexity apparently wouldn’t agree.

After the deal collapsed, CNN sent Perplexity a cease-and-desist letter demanding it stop using CNN’s content and trademarks. Perplexity allegedly never responded. Now they’re in federal court.

Nine Publishers and Counting

CNN is joining a queue. The full list of entities that have sued Perplexity for copyright infringement:

Nine plaintiffs. One defendant. At some point, this stops looking like a string of isolated disputes and starts looking like a business model problem.

The Anthropic Benchmark

The one settlement that matters: in August 2025, Anthropic agreed to pay $1.5 billion to resolve a class-action lawsuit from a group of authors led by writer Bartz. A fairness hearing was held on May 14, 2026, and final approval is pending.

That number — $1.5 billion — is now the benchmark every publisher has in mind when they sit down at the negotiating table. It’s also the number AI companies are desperate to avoid setting as precedent. If the Bartz settlement gets final approval, every subsequent lawsuit gets harder to settle cheaply.

Perplexity’s Defence: “You Can’t Copyright Facts”

Perplexity’s Chief Communications Officer Jesse Dwyer gave a one-line response to CNN: “You can’t copyright facts.”

It’s a clever line. It’s also a legal position with real precedent behind it. The 1991 Supreme Court case Feist Publications v. Rural Telephone Service Co. established that facts alone cannot be copyrighted — only the original expression of those facts.

The problem for Perplexity is that CNN isn’t suing over facts. They’re suing over:

• Verbatim article text — which is original expression, not facts
• Photos and videos — which are creative works with their own copyright
• Paywalled content — which was reproduced without authorization or compensation
• Hallucinated attributions — fabricated content falsely branded as CNN reporting

The “facts” defence works if Perplexity was extracting data points and synthesizing new answers. It doesn’t work if — as CNN alleges — Perplexity was reproducing substantial portions of original reporting verbatim and serving it up as its own output.

The Other Path: Licensing Deals

Not every publisher is suing. Several have chosen the cheque:

• Time signed a licensing deal with Perplexity
• Gannett reached an agreement
• Le Monde partnered rather than sued
• Multiple outlets have signed deals with OpenAI and Google

The split is becoming clear: publishers with enough leverage to sue (NYT, News Corp, CNN) are going to court. Publishers with less leverage are cutting deals. The ones in the middle — regional newspapers, specialist outlets — are getting scraped with no compensation and no recourse.

What Happens If CNN Wins

If the court finds that Perplexity’s reproduction of verbatim article text violates copyright, it establishes that AI output which substantially reproduces source material isn’t fair use. That would force every AI search company to either:

1. License the content they surface, or
2. Build systems that genuinely synthesize rather than regurgitate

That’s a meaningful distinction. An AI that reads 50 articles about a topic and writes a genuinely new summary is different from an AI that reads one article and reproduces large chunks of it.

What Happens If Perplexity Wins

If the court sides with Perplexity — particularly on the “facts can’t be copyrighted” argument — it essentially greenlights AI companies to harvest and redistribute journalistic content as long as they reframe it as “answering questions” rather than “publishing articles.” That would be catastrophic for newsroom economics. Why subscribe to CNN when Perplexity will give you the same reporting for free, attributed to “AI search results”?

❓ Frequently Asked Questions

Q: What does this mean for NZ media? NZ news outlets like RNZ, NZ Herald, and Stuff haven’t sued any AI company yet — but they’re watching. NZ copyright law has similar fair-dealing provisions to US fair use. A CNN win would strengthen the hand of any NZ publisher considering action against AI scrapers.

Q: Is Perplexity the only AI company being sued? No. OpenAI faces its own lawsuits from NYT and other publishers. The difference is scale — Perplexity has attracted more lawsuits faster, partly because its core product (AI-powered search answers) directly surfaces content rather than just training on it.

Q: Should I stop using Perplexity? That’s a personal call. The legal question isn’t about users — it’s about the company’s practices. But if you care about publishers getting paid for their work, it’s worth considering whether the AI tool you use compensates the sources it draws from.

🔍 THE BOTTOM LINE

Nine publishers suing one AI company isn’t a coincidence — it’s a pattern. Perplexity built a product that consumes journalism and excretes answers. The courts now have to decide whether that’s fair use or theft. The answer won’t just affect Perplexity. It’ll shape the entire relationship between AI and the media for the next decade.

SOURCES

• CNN
• The Verge
• Al Jazeera
• Engadget
• Bloomberg Law
• Courtlistener (Docket 73402641)