The New York Times has filed a high-stakes lawsuit against Perplexity AI, accusing the startup of unlawfully copying, distributing and displaying millions of its articles to train and power its generative AI products. The complaint alleges that Perplexity’s core business model is built on scraping protected content, including material behind paywalls, without licence or compensation.
The lawsuit, filed in the US District Court for the Southern District of New York, states that The New York Times’ journalism was used to create AI-generated summaries, responses and explanations that were served to Perplexity users as original content. According to the Times, the platform also produced fabricated outputs or “hallucinations” that were falsely attributed to the newspaper and displayed alongside its registered trademarks, misleading consumers and damaging brand credibility.
The Times said it supports responsible AI development but firmly rejects the unauthorized use of its content to build commercial products. The company is seeking damages, injunctive relief and orders that would prevent Perplexity from continuing what it describes as widespread copyright infringement.
Perplexity has dismissed the allegations, calling lawsuits from publishers an outdated strategy aimed at resisting new technology. The company previously stated that it does not scrape data to train large foundation models, but instead indexes publicly available web pages and offers factual citations to users.
The legal case reflects escalating tensions between media companies and AI developers over the use of copyrighted material. Publishers argue that journalism requires costly labour and editorial resources, and that AI firms cannot bypass licensing systems while monetizing their outputs. AI companies counter that web content remains accessible for fair indexing, particularly where it is not clearly restricted.
The dispute also illustrates the broader struggle over who controls and profits from digital information. The Times has existing agreements with firms such as Amazon, which uses its content for AI-powered products like Alexa. However, it has not reached similar arrangements with Perplexity or other emerging competitors.
Perplexity, valued at nearly $20 billion, is facing a growing list of lawsuits. Recent complaints have been filed by the Chicago Tribune, Encyclopedia Britannica, and media outlets owned by Rupert Murdoch, including Dow Jones and the New York Post. The startup is also fighting a separate lawsuit brought by Reddit, which accuses AI companies of scraping user-generated data without authorization.
The Times previously sent Perplexity a cease and desist notice in 2023, but claims the startup continued to utilise its content. The lawsuit represents another flashpoint in the rapidly expanding debate over data ownership, digital rights and the future economics of AI-powered information systems.
