The New York Times has spent more than $20 million on its copyright lawsuit against OpenAI and Microsoft, and publisher A.G. Sulzberger has no intention of settling. In an extended interview with Wired, Sulzberger laid out the case that the legal fight is not just about one newspaper’s revenues – it is about whether the economic foundation of original journalism can survive the AI era.
The lawsuit, filed in December 2023, alleges that OpenAI and Microsoft used millions of copyrighted Times articles to train their language models without permission or compensation. The case has since expanded to include Perplexity, an AI search startup, and has become the highest-profile test of whether training AI on copyrighted material constitutes fair use or infringement. A ruling is not expected before 2027.
Sulzberger’s argument, detailed in both the Wired interview and a speech at the WAN-IFRA World News Media Congress in June, is that the AI industry has built its business model on top of the collective output of human creators without paying for it. He argues that the problem extends beyond journalism to books, film, music, and academic research – creative industries that collectively employ more than 50 million people worldwide and generate roughly $12 trillion in annual economic value.
The Times has taken a notably harder line than some of its peers. Several major publishers have signed licensing deals with AI companies, accepting payments in exchange for the right to train on their archives. Sulzberger argues that these deals, while individually rational, collectively create a race to the bottom. If every publisher accepts whatever price an AI company offers, the total compensation will never approach the value of the content being consumed.
The risk for the Times is existential in a different way. AI-generated summaries in search results and chatbot responses reduce the incentive for readers to click through to news articles. If traffic continues to decline, the advertising and subscription revenue that funds the Times’ newsroom – one of the largest in the world – will follow. And the legal remedy, even if the Times wins, may arrive too late to reverse the structural damage.
The case has already survived early motions to dismiss, and discovery is underway. OpenAI has argued that training on publicly available text falls within fair use, a position supported by some legal scholars but challenged by the Times’ argument that the models can reproduce Times articles verbatim, competing directly with the original.
For smaller news organizations watching from the sidelines, the outcome matters as much as the strategy. The Times can afford a $20 million legal bill. Most local newspapers cannot. If the fair-use question is resolved in OpenAI’s favor, small publishers will have little legal recourse and even less negotiating leverage. If it is resolved against OpenAI, the damages framework will determine whether AI companies pay enough to sustain the journalism they depend on.
Sources: Can the New York Times Save Journalism From Our AI Overlords? (Wired, July 28, 2026); New York Times publisher A. G. Sulzberger on why news publishers should fight AI platforms (Reuters Institute, June 1, 2026); The New York Times sues OpenAI and Microsoft for copyright infringement (CNN, December 27, 2023)