US Justice Department files statement of interest backing OpenAI and Microsoft in the New York Times copyright case

What happened

On 1 September 2026, the United States filed a statement of interest in the consolidated OpenAI copyright litigation (the case brought by The New York Times against OpenAI and Microsoft, filed in late 2023). In a 20-page filing, the Justice Department argued that training large language models on copyrighted works is "exceedingly transformative" and can qualify as fair use, and that the creative possibilities and public benefits of LLM training "far outweigh any competitive harm." It argued that constraining LLM development under a different reading of fair use would thwart creative and scientific progress, and that the Times' interpretation would create an anticompetitive entry barrier functioning as large subsidies for established media companies. Associate Attorney General Stanley Woodward publicly linked the position to the administration's view that AI dominance is critical to national security and prosperity. A New York Times spokesman responded that the administration was siding with trillion-dollar AI companies at the expense of American creators, and that letting companies take content without permission or compensation would undermine the sustainability of human-created content.

Why it matters to creators

This is the US government formally weighing in, on the AI companies' side, in the central fair-use fight over training on copyrighted journalism. A statement of interest is advocacy, not a court ruling, but it signals how the executive branch wants courts to treat unlicensed training — directly affecting whether authors and publishers can expect permission and payment requirements for training use of their work.

What is not yet known

The court has not ruled on the fair-use questions in the consolidated litigation, and the statement of interest does not bind the judge. How much weight the court gives the government's position, and publishers' counter-arguments (including that licensing costs are small relative to AI companies' infrastructure spending), remains to be seen.