The legal battles between AI companies and traditional media just got messier. Last week, The New York Times filed a lawsuit against Perplexity AI, the buzzy search startup backed by Amazon (AMZN) founder Jeff Bezos, accusing it of systematically stealing journalism to power its generative AI tools.

The Core Allegations: Scraping Without Permission

Filed Friday in the U.S. District Court for the Southern District of New York, the lawsuit claims Perplexity built its business on the backs of content it never paid for. According to The Times, the company copied, distributed, and displayed millions of articles, including stories behind paywalls, without authorization or licensing agreements.

Graham James, a spokesperson for The Times, put it bluntly: the content Perplexity used to train and advance its products "should only be accessible to our paying subscribers." Translation: if you want our journalism, you need to pay for it like everyone else.

The lawsuit comes more than a year after The Times issued a cease-and-desist notice to Perplexity, which apparently didn't resolve the dispute.

Hallucinated Content With NYT Branding

But the allegations go beyond simple copying. The Times also claims Perplexity's AI models generated false or misleading summaries and then slapped The Times' branding on them, making it look like the newspaper had reported information it never actually published.

These so-called "hallucinations" aren't just embarrassing technical glitches. The filing argues they pose real reputational risks and confuse readers about where information actually comes from. When an AI tool puts your logo next to fabricated content, that's a problem.

Perplexity Pushes Back

Perplexity AI isn't backing down. Jesse Dwyer, the company's communications chief, dismissed the lawsuit as a tired playbook publishers use whenever new technology threatens their business model, according to Reuters.

The company maintains it doesn't scrape data to train foundation models. Instead, Perplexity says it indexes web pages and provides factual citations, more like a search engine than a content thief.

Perplexity, which is currently valued at roughly $20 billion, didn't immediately respond to requests for additional comment.

A Broader Pattern Emerges

The Times isn't alone in taking legal action. The Chicago Tribune also filed a lawsuit against Perplexity on Thursday, suggesting this could become a coordinated effort by traditional publishers to push back against AI companies they believe are profiting from their work without compensation.

The outcome of these cases could set important precedents for how AI companies can legally use content from news organizations and other publishers.