The New York Times announced a lawsuit Friday against AI search startup Perplexity for copyright infringement, its second such case against an AI firm. The Times is one of several news organizations suing Perplexity, among them the Chicago Tribune, which also filed suit this week.
The Times’ lawsuit says that Perplexity “offers commercial products to its own users as alternatives” to it, “without permission or remuneration.”
The lawsuit — one of the latest tactics in the years-long campaign, now taking place even as a handful of publishers, including The Times, cut deals with AI companies — is part and parcel of that same, protracted game plan. Underneath this is the understanding that a tsunami of AI can’t be stopped, but also an effort to use lawsuits as leverage in negotiations and to force AI companies to legally license content in ways that reward creators and keep original journalism economically viable.
Perplexity attempted to address gripes about compensation by debuting a Publishers’ Program last year, which provides participating outlets such as Gannett, TIME, Fortune, and the Los Angeles Times with a share of ad revenue. In August, Perplexity also introduced Comet Plus, which gives 80% of its $5 monthly fee to participating publishers, and it recently inked a multi-year licensing deal with Getty Images.
“While we believe in the ethical and responsible use and development of AI, we vigorously object to Perplexity’s unauthorized access to our content for the purpose of developing and promoting their products,” Graham James, a Times spokesman, said in a statement. “We will keep fighting to hold companies accountable when they try to pretend our work isn’t of value.”
Like the Tribune’s suit, the Times is concerned about Perplexity’s approach to answering users’ questions — by pulling information from other sites and databases to generate responses via its retrieval-augmented generation (RAG) products, including chatbots and Comet, its browser AI assistant.
“Perplexity then republishes the original content in written responses to users,” the suit says. “Those responses, or outputs, are frequently verbatim or near-verbatim reproductions, summaries, or abridgments of those original works — including The Times’s copyrighted articles.”
Or, as James said in the statement: “RAG permits Perplexity to scrape the web and pirate content from behind our paywall and feed it to its subscribers instantaneously. That is material that should be available only to our paying subscribers.”
The Times also alleges that the search engine of Perplexity has produced hallucinated information and falsely attributed it to the platform, which negatively impacts its brand.
“Publishers have sued every new tech format for the last one hundred years, including radio, TV, internet, and social media, and now AI,” Perplexity head of communications Jesse Dwyer said in a statement to TechCrunch. “Thankfully, it’s never worked, or we’d still be discussing this by telegraph.
(In the past, publishers have also sometimes won or influenced major legal fights over new technologies, and from these emerged settlements, licensing regimes, and court rulings.)
The lawsuit was filed a little over a year after The Times sent Perplexity a cease-and-desist letter in response to its summaries and other output using content from the paper. According to the outlet, it has been in touch with Perplexity on multiple occasions over the past year and a half, requesting that it stop using its materials unless a licensing agreement could be struck.
For The Times, it is not the first battle it has picked with an AI company. The Times is also suing OpenAI and its sponsor, Microsoft, alleging that the two entities trained their AI models on millions of articles published by the outlet without compensation. OpenAI has claimed that training AI on public data is “fair use” and has fired back at the Times, alleging that the magazine manipulated ChatGPT to search for evidence.
That case is ongoing, but a similar lawsuit against OpenAI competitor Anthropic could help set a precedent on fair use for training AI systems. There, in a case in which authors and publishers sued the AI company for using pirated books to train its models, the court found that while obtaining the books lawfully might offer a secure fair use defense, unauthorized use is a copyright violation. Anthropic settled for $1.5 billion.
The Times’ suit is the latest in a slew of lawsuits aimed at pressuring Perplexity. Last year, News Corp’s owners of publications such as The Wall Street Journal, Barron’s, and the New York Post made similar allegations against Perplexity. That list expanded in 2025 to also feature Encyclopedia Britannica, Merriam-Webster, Nikkei, Asahi Shimbun, and Reddit.
Other news agencies, like Wired and Forbes, have labeled Perplexity a “scrapper” and accused the startup of stealing and of unethically crawling websites that have specifically said they don’t want to be scraped. The latter assertion has recently been verified by the internet infrastructure provider Cloudflare.
In its lawsuit, The Times is seeking to have the courts order Perplexity to compensate it for the harm it claims was caused and to prevent the startup from further using its content.
The Times is not above partnering with AI companies that pay for its reporters’ work, clearly. The outlet earlier this year signed a multi-year agreement with Amazon to license its content to help train the tech giant’s AI models. AI firms have inked licensing deals with several other publishers and media companies to use their content for training processes and chatbot responses. OpenAI has partnerships with the Associated Press, Axel Springer, Vox Media, The Atlantic, and other partners.
