On Thursday, several news organizations accused OpenAI of withholding evidence about how the company trains its artificial intelligence models in a new move linked to a series of ongoing copyright lawsuits.
The proposal was filed by 17 publishers, including the New York Times, the New York Daily News, the Chicago Tribune and Ziff Davis (CNET’s parent company). Ziff Davis sued OpenAI in 2025, alleging that OpenAI infringed on its copyrighted works to train ChatGPT and other major languages.
The first case dates back to 2023 when the New York Times began suing OpenAI and Microsoft, alleging that the companies were building their AI technology using millions of news articles written by journalists. Microsoft and OpenAI have denied the claims.
The motion asks the court to impose legal sanctions against OpenAI, but not Microsoft, for allegedly withholding evidence, such as datasets and output logs, and says “OpenAI chose to be restrained” by failing to disclose it. If those sanctions are imposed, OpenAI could be ordered to pay financial penalties.
“This motion asks the court to punish OpenAI for concealing and destroying evidence showing how ChatGPT was trained on stolen journalism,” New York Daily News attorney Steven Lieberman said, according to the Associated Press.
At the heart of the case is how generative AI, such as ChatGPT, is trained and how it acquires its knowledge. The Times’ first lawsuit says OpenAI’s AI-generated tools “can produce output that replicates Times content verbatim, closely summarizes it, and mimics its expressive style,” raising questions of copyright infringement.
The lawsuits come amid a broader conversation in the journalism industry: declining traffic across digital media outlets. The overview of AI is often cited as a major reason for the decline in clicks on original reports by authors and editors, which in turn impacts publishers’ advertising revenue.
The growing reliance on AI chatbots to find news and other content is also a big problem for publishers, as it eliminates genuine reading and audience engagement. Some data shows that small publishers have been hit hardest, with a reported 60% drop in traffic, while other analyzes predict a traffic drop of over 40% by 2029.
A statement from Ziff Davis notes that “OpenAI has copied and monetized Ziff Davis content without permission on a large scale.” Lance Koonce, a partner at Klaris Law and counsel for Ziff Davis, said that, since the lawsuit, “OpenAI repeatedly lied about its ability to search its data sets for Ziff Davis content and engaged in other criminal misconduct.”
The ongoing debate about copyright and AI
OpenAI has long maintained that AI training is a worthy use. An OpenAI spokesperson denied the allegations in a statement sent to CNET, saying: “As the Times’ case weakens and they are forced to withdraw their claims against us, they continue their efforts to invade the privacy of people who have nothing to do with this case, including making these allegations blatantly false.” The statement continued: “We will continue to protect our users’ privacy and long-established principles of fair use.”
In a 2024 rebuttal to the first lawsuit filed by the New York Times, OpenAI said the publisher falsely accused the company of destroying data and instead accused the newspaper of “secretly” removing its data that would have shown internal use of OpenAI products. Although the Times has dropped one claim against OpenAI, the main case is still in court.
Other tech giants, incl Metathey were also sued by authors and news publishers for copyright infringement. Many of those cases are still pending as courts decide where to draw the line between fair use and infringement in the age of AI.
