US backs OpenAI in NYT copyright fight

On Sept. 2, 2026, the U.S. Department of Justice filed a brief in Manhattan federal court supporting OpenAI in a copyright lawsuit brought by The New York Times (NYT) and several other newspapers. The case centers on whether training large language models on copyrighted texts infringes the rights of publishers and creators. The government’s filing sets out the administration’s legal and policy view on how copyright law should apply to the development of advanced AI systems.

The Justice Department brief describes the process of training large language models on copyrighted texts as "extraordinarily" transformative. It states that such training generally makes fair use of copyrighted material and urges the court to reject arguments that this practice violates copyright law. The filing says the United States has a strong interest in ensuring the court does not adopt a reading of copyright that would treat AI training on such texts as infringement.

The brief warns that constraining LLM development under what it calls a misunderstanding of fair use would "thwart such creative and scientific progress." It further argues that restrictive interpretations could harm national security, U.S. economic competitiveness, and broader American prosperity. The administration thus links the outcome of the case to the country’s ability to maintain leadership in AI research and deployment.

The New York Times (NYT) publicly criticized the government’s intervention and its support for OpenAI. It said the administration is siding with "a handful of trillion-dollar AI companies" at the expense of "countless American creators whose work they stole." The newspaper argues that AI companies should "pay fairly for the content" that underpins their products, framing the dispute as one about compensation and rights for creators whose works are used in training.

This response highlights a sharp divide between the administration’s focus on innovation and competitiveness and publishers’ emphasis on protecting creative output. For the Times and other plaintiffs, the core issue is whether the unlicensed use of copyrighted material in AI training should trigger payment or other protections for rights holders.

Associate Attorney General Stanley Woodward Jr. publicly defended the administration’s position on social media platform X. He said the administration would not allow the United States to be disadvantaged relative to foreign adversaries by what he called a "plainly incorrect understanding of copyright law." His comments emphasize the government’s view that a stricter interpretation of copyright in this context could weaken the U.S. in global AI competition.

The Justice Department filing is advisory and does not decide the underlying litigation, which remains before the Southern District of New York. The court will ultimately determine how copyright law applies to the training of large language models, but the brief signals how the federal government believes that balance should be struck between fostering AI development and enforcing copyright protections.