USA Today wants over $250 million from OpenAI

"OpenAI's commercial success rests on large-scale copyright infringement," reads the lawsuit USA Today Co. filed Thursday in Manhattan federal court, Reuters reports. The publisher and several local papers it owns are asking for more than $250 million in damages.

At a glance

• USA Today Co. sued alongside local papers it owns, including The Tennessean, Indy Star, The Columbus Dispatch and The Oklahoman, in the Southern District of New York under case number 1:26-cv-08892.
• The complaint says OpenAI copied "hundreds of thousands" of articles to train its language models without asking, and claims damages above $250 million for "real and continuing" harm.
• The catch: the closest parallel, The New York Times case against OpenAI, still has no verified trial date, so nobody yet knows how a court will answer the core fair-use question.

If you have not been following, The New York Times was the first major news organization to sue OpenAI and Microsoft, according to Wikipedia, which dates that filing to December 27, 2023. The Times case was consolidated with related suits on April 3, 2025, and now sits in multidistrict litigation 1:25-md-03143, where local publishers such as Chicago Tribune and Orlando Sentinel entities are also filing.

USA Today puts its harm above $250 million and brings in at least four local papers

The suit names USA Today Co. and several local newspapers it owns, among them The Tennessean, Indy Star, The Columbus Dispatch and The Oklahoman. It was filed in the US District Court for the Southern District of New York as case 1:26-cv-08892. Steven Lieberman of Rothwell Figg Ernst & Manbeck represents the publishers.

The core allegation is about volume. According to the complaint, OpenAI copied "hundreds of thousands" of articles from USA Today and its affiliated papers to train its large language models and never asked for permission. The publishers say OpenAI "took" that content to "build products worth hundreds of billions of dollars" instead of licensing it.

The filing describes the damage to the outlets as "real and continuing." The publishers also argue that reporting, writing and editing take serious investment, and that OpenAI used the results of that work without paying for it or crediting it.

USA Today joins The New York Times and nearly 400 local newspapers already suing OpenAI

The list of plaintiffs already lined up against OpenAI is long. It includes The New York Times, The Intercept, CNET owner Ziff Davis, CBC/Radio-Canada, Encyclopaedia Britannica, Merriam-Webster and The Seattle Times, plus a coalition of nearly 400 local newspapers.

The Times case is the furthest along. According to Legalaiinsights, Judge Stein's March 26, 2025 ruling on the motion to dismiss let the direct and contributory infringement claims survive, dismissed the DMCA §1202 claims with leave to amend and threw out common-law unfair competition with prejudice. That ruling did not decide whether infringement happened.

Summary-judgment briefing in the consolidated case closed around April 2, 2026, the publishers filed a sanctions motion on July 9, 2026, and no trial date has been verified. According to AI Lawsuit Tracker, Magistrate Judge Wang ordered production of a de-identified sample of 20 million ChatGPT logs, and Judge Stein affirmed that order on January 5, 2026.

How does a fair-use defense meet a damages claim?

OpenAI's answer in the Times case is fair use, the US doctrine that permits some unlicensed use of copyrighted work, weighed partly by whether the new use replaces the original in the market. According to Wikipedia, OpenAI argued that training on copyrighted material is fair use and that its products did not serve as a market substitute for the Times.

According to AI Lawsuit Tracker, a Sept. 2, 2026 Justice Department statement of interest backed OpenAI's position that barring AI training on copyrighted material could hamper technological progress. The same tracker notes that statutory damages for willful infringement run up to $150,000 per work, which is why exposure in these cases is framed as potentially existential.

Statutory damages work like a fine charged per item on a receipt rather than per shopping trip, so a case built on many works can reach large sums quickly. Licensing is the other route. According to Wikipedia, AP signed a deal with OpenAI in July 2023 and Axel Springer followed in December 2023, and The Information valued some publisher agreements at $1 million to $5 million a year.

The filing, as reported, does not say which OpenAI models were trained on the articles, how the publishers arrived at "hundreds of thousands," or whether the case will join the consolidated docket where other newspaper suits sit. In our view, the price gap is hard to ignore: the $250 million claim sits far above the $1 million to $5 million a year that some licensing deals were reportedly valued at.

What OpenAI files in 1:26-cv-08892

OpenAI's formal answer to the complaint is the next checkpoint, and the reports give no deadline for it. Two questions remain open. One is whether the case gets folded into multidistrict litigation 1:25-md-03143 alongside the Times and other newspaper suits. The other is whether a summary-judgment ruling there, with briefing closed since around April 2, 2026, arrives first and shapes how this one is argued.