News outlets including The New York Times and the Daily News asked a Manhattan federal judge to sanction OpenAI, accusing the company of discovery misconduct in the copyright case over how ChatGPT was trained on news content. OpenAI denies wrongdoing and says it is protecting user privacy and relying on fair use.
The New York Times, the Daily News and other news organizations have asked a federal judge in Manhattan to sanction OpenAI, escalating the copyright fight over how ChatGPT was trained on news articles.
The plaintiffs said in a motion filed Thursday, July 9, 2026, that OpenAI engaged in discovery misconduct by withholding datasets and ChatGPT logs that could show how copyrighted news content was used. They also said a recent deposition of an OpenAI employee contradicted earlier claims.
OpenAI spokesperson Drew Pusateri rejected the allegations, saying they were false and that the company was defending user privacy while continuing to rely on fair use.
How the dispute reached this point
The sanctions request is the latest turn in a lawsuit the Times brought against OpenAI and Microsoft in late 2023. That case is part of a wider wave of copyright challenges against AI companies over the use of online material to train large language models.
New York Daily News attorney Steven Lieberman said OpenAI has spent two years making misrepresentations about its ability to search copyrighted content in training datasets and logs.
The plaintiffs say the disputed records matter because they could help show exactly how OpenAI’s systems handled copyrighted news material. The company, by contrast, says the plaintiffs are trying to override privacy protections.
What is at stake
The dispute adds another layer of pressure to a case with significant financial and legal stakes for both publishers and AI companies. A sanctions ruling could bring attorney-fee penalties or other consequences if the judge agrees OpenAI mishandled discovery.
More broadly, the case could shape what evidence is available before a landmark copyright trial and influence how courts treat AI training on news content under U.S. copyright law.
The next procedural step is the judge’s response to the sanctions motion, along with any opposition filings from OpenAI or further court orders on the disputed datasets and logs.
Revision note
Initial automated publication.