Justice Department backs OpenAI in copyright fight with The New York Times

The Justice Department is backing OpenAI in its copyright fight with The New York Times, warning that a ruling against the artificial intelligence company could reach beyond copyright law.

In a statement of interest filed in the consolidated OpenAI copyright litigation and reported on by Nieman Labs, the government argued that finding OpenAI’s training practices outside the bounds of fair use could “threaten national security” and give “a competitive advantage to foreign adversaries.”

The case combines claims from the Times with lawsuits brought by other publishers, book authors and nonprofit news organizations, including The Intercept. The Justice Department said it intervened because it wants American AI companies to remain “competitive” and able to “develop a robust AI industry.”

At the center of the dispute is fair use. OpenAI and other AI companies argue that training a large language model on copyrighted material is transformative because the resulting model serves a different purpose from the material used to train it.

The government largely embraces that argument. A Times article is meant to inform or educate a reader, the filing says, while an AI model can perform tasks such as writing, translating and solving problems.

The Times has also accused ChatGPT of reproducing some of its articles. The Justice Department acknowledged that claim but argued that the case concerns training rather than outputs, and that courts should not focus on unusual or hypothetical uses of the product.

The filing also makes a case around smaller publishers. The government argues that underresourced newsrooms can use AI tools “to compete,” including by generating an image that might otherwise require a photographer or license. Requiring AI developers to pay licensing fees across large archives, it says, could disproportionately benefit major publishers.

The plaintiffs sharply disagree.

“The Administration is siding with a handful of trillion-dollar companies at the expense of the countless American creators whose work they stole,” Times spokesperson Graham James said.

Matt Topic, an attorney representing The Intercept, called the government’s position an “unprecedented, uncompensated transfer of IP rights from news organizations to tech companies.”

The Justice Department also pointed to AI use at the Times, citing a Futurism report saying Times authors use LLMs to “conceptualize and edit” articles. But Nieman Lab noted that the phrase referred to a freelance Modern Love contributor who used AI without the disclosure the newspaper requires. The Times has separately cut ties with a freelancer over AI-assisted work and is testing AI summaries built only from its own journalism.

The filing puts the federal government on OpenAI’s side in one of publishing’s most closely watched copyright fights. That follows the administration’s earlier AI policy framework, which sided with AI developers on copyright.

For smaller publishers, the stakes are less simple. The Tow Center maintains a tracker of AI deals and lawsuits, while Nieman Lab notes that small publishers have seen little licensing revenue from OpenAI. Litigation offers another possible route, but its cost means few small and nonprofit news organizations have sued.

Favorable legal precedent, Nieman Lab notes, could give those publishers more leverage and potentially create new revenue streams.

“The principle of being able to take content without compensation is critically important to us — and so is our survivability,” The Intercept general counsel David Barlow told Nieman Lab in 2024. “They go hand in hand.”