DOJ Sides With OpenAI, Says AI Training on Copyrighted Works Is Fair Use

The US Department of Justice has filed a court submission in support of OpenAI in a copyright lawsuit brought by The New York Times, saying that training artificial intelligence on copyrighted material is fair use. The position was reported on September 2, 2026, by Wired and TechCrunch, both of which described the filing as an intervention by the Trump administration on behalf of a leading AI company. Wired said the administration is now siding with OpenAI in the New York Times suit. TechCrunch characterized the filing as a government defense of the practice of training large language models on copyrighted content. In its filing, the government warns that restricting such training could "thwart creative and scientific progress," a phrase quoted by Wired.

TechCrunch quoted the filing asserting that the United States has a strong interest in continuing to build a robust and competitive AI industry and in shaping global practice and procedure for AI use. That language indicates that Washington views the case as more than a private dispute between a content owner and a technology developer, but rather as a matter of industrial policy. The underlying litigation stems from the mechanics of generative AI. Large language models learn by scanning vast quantities of text drawn from books, newspapers and websites, including material under copyright. The New York Times and other copyright owners argue that this use without permission is unlawful. OpenAI, backed by the DOJ, says it constitutes fair use.

The court is thus being asked to define the limits of a doctrine that has operated for centuries but is now being tested by machine learning. Fair use, a cornerstone of American copyright law, permits unauthorized use of protected material in certain circumstances. It requires a case by case evaluation of factors such as the purpose of the use, the nature of the original work, the amount used, and the effect on the market for that work. The present dispute focuses on whether ingesting copyrighted text to train a model is a legitimate, transformative use or an infringement that strips works of their economic value.

The government has not provided the final answer, but it has urged the court to account for the broader consequences of limiting AI training. The DOJ's filing is not binding on the judge. Nevertheless, courts have often treated executive branch views with respect when cases touch on national policy and the governing law is vague. For other AI developers facing similar lawsuits, the government's position offers support: their training practices now align with what Washington considers competitive priorities. For media companies, it is a sign that copyright lawsuits over training data must overcome the declared stance of the federal government. Should the court rule that training on copyrighted material is fair use, OpenAI and its peers would avoid licensing payments for such data.

If the court rejects that claim, they would be forced to license content or restrict themselves to public domain works, increasing the cost of building models and reshaping the industry's approach. The stakes therefore extend to the cost structure of AI and the future balance of power between technology companies and content creators. By joining the case, the DOJ has turned a private copyright battle into a national interest question. Wired framed the move as the Trump administration embracing an AI giant. In its content, the filing reveals a broader calculation: the United States wants its AI industry to survive and lead, and it wants American practice to be the standard for the world. The word "global" is potent.

How the US courts resolve this issue could serve as a reference point for other jurisdictions where similar questions are being litigated. Both Wired and TechCrunch covered the development on the same day, and the litigation is continuing. The judge will weigh the government's argument beside the parties' briefs. What has undeniably changed is the role of the state. The federal government, which might have stayed silent, now stands behind OpenAI. For copyright holders, the message is unmistakable: the executive branch supports the AI industry's interpretation of fair use, and attempts to extract compensation through litigation will face that official position. Whether the judiciary agrees with the executive branch will be decided in due course.