According to PYMNTS.com, investigative journalist John Carreyrou, author of “Bad Blood,” filed a copyright lawsuit on December 10 against six major AI companies: Google, Meta, OpenAI, Anthropic, Perplexity, and xAI. The suit alleges these firms used his book without permission to train their AI models. Carreyrou and his co-plaintiffs specifically chose not to join a broader class action, arguing such suits allow defendants to settle many cases at “bargain-basement rates.” This news follows The New York Times suing Perplexity on December 5 for copyright infringement, and a report that Disney sent a cease-and-desist to Google over its AI tools on December 7. In a related case, a judge ruled in June that Anthropic may have illegally downloaded as many as 7 million books.
Why This Case Stands Out
Here’s the thing: this isn’t just another faceless class action. Carreyrou is a specific, high-profile journalist with a massive, definitive work. He’s not just claiming his words were scraped; he’s arguing the very value of his investigative work—the facts he uncovered about Theranos—was ingested to make these AI models smarter. And his strategy of going it alone is fascinating. He’s basically saying, “Class actions let you guys off too easy.” He wants a real fight, not a cheap settlement that gets folded into some giant pool. That makes this a much bigger headache for the defendants.
The Perplexity Problem
Now, Perplexity is really in the hot seat. They’re facing the Carreyrou suit *and* the NY Times suit, which is brutal for a startup. The Times’ accusation is particularly damning: not just copying, but sometimes making up info and falsely saying it came from The Times. That hits their core value proposition as a reliable answer engine. Their spokesperson’s response about publishers suing new tech for a hundred years? It’s a classic deflection. But this time, the plaintiffs have very specific, very angry lawyers. And Perplexity’s claim they “don’t index books” feels like a narrow, technical dodge in a much bigger war about where training data comes from.
A Industry-Wide Reckoning
So what’s the endgame here? Look at the pattern: Anthropic already settled with authors in August. Disney is firing warning shots at Google. The legal pressure is coming from all sides—individual authors, giant media corporations, and now investigative reporters. The AI companies’ “move fast and break things” approach to data is facing its ultimate test. I think we’re going to see a messy, expensive, and prolonged legal brawl that fundamentally changes how models are trained. Will it be licensing deals? Will it be massive fines? Or will the courts surprisingly side with the tech giants? One thing’s for sure: the free-for-all era of data scraping is over. The bill is coming due.
