DOJ Backs OpenAI's Fair-Use Case in NYT Copyright Suit

Introduction

On September 1, 2026, the U.S. Department of Justice filed a Statement of Interest in the consolidated copyright litigation against OpenAI in the U.S. District Court for the Southern District of New York, before Judge Sidney H. Stein. The roughly 20-page filing is not an amicus brief. Several news outlets referred to it as a "brief," but DOJ's own filing is titled a Statement of Interest — the specific legal vehicle federal agencies use to state the government's position in private litigation without becoming a party to the case. It was signed by associate attorneys general Stanley Woodward Jr. and Brett Shumate, along with attorney general's senior counsel Michael Weisbuch. The filing became widely reported on September 2, 2026.

The case is the consolidated action brought by The New York Times, The Intercept, Tribune Media Company, and Ziff Davis. The Intercept's own suit, filed against OpenAI and Microsoft in February 2024, was narrowed in November 2024 when the then-presiding judge dismissed all claims against Microsoft while allowing the DMCA claim against OpenAI to proceed; that case was later consolidated under Judge Stein alongside the Times' suit and others. The filing is reported as the first time the U.S. government has intervened directly in the wave of AI training-data copyright suits, according to coverage of the filing; DOJ's own filing does not make that claim.

Filing Overview

DOJ's central argument is that training large language models on copyrighted written works generally qualifies as fair use. The filing contends that LLMs do not reproduce the text they train on; instead, training builds generalized reasoning and language capabilities through a process the filing calls "extraordinarily transformative." DOJ accused the Times and the other plaintiffs of pressing an overly narrow reading of fair-use doctrine: "In this litigation, the New York Times seeks to narrow fair-use doctrine to exclude the training of OpenAI's large language models (LLMs). That result would be inconsistent with basic copyright law principles and severely hamper 'the Progress of Science and useful Arts.'"

The filing frames the stakes in market-structure terms as well as doctrinal ones, warning: "An erroneous fair use ruling would hamper competition in the market for LLMs, because only the largest technology companies might have the capital necessary to pay licensing fees." DOJ also ties the outcome to innovation broadly: "Constraining LLM development under a misunderstanding of fair use doctrine would thwart such creative and scientific progress." Beyond copyright doctrine, the filing argues that AI development is critical to national security, and situates the case within a stated federal interest: "The United States has a strong interest in continuing to develop a robust and competitive artificial intelligence industry that sets the standard for the practice and procedure of AI use globally."

Legal Significance and What Happens Next

A Statement of Interest is not a ruling, and it does not bind the court. Judge Stein decides the fair-use question independently, weighing DOJ's filing alongside the parties' own summary judgment briefing and the existing record. Its practical effect is to add the government's stated position to the docket, without making DOJ a party to the case or subjecting it to discovery obligations.

Timing sharpens the story. Expert discovery in the case closed August 3, 2026, and the court had already ordered summary judgment and Daubert motions due September 4, 2026 — just three days after DOJ filed.

Matt Topic, the attorney litigating on behalf of The Intercept, disputed the filing's basis. He noted that DOJ "has not been privy to any discovery in the case," adding: "What they're saying isn't based on any understanding of actual facts." Topic also characterized the intervention as political: "This is pretty reflective of the administration's pro-AI policies, which are out of step with the public skepticism of AI and out of step with our long tradition in this country of protecting human expression."

Implications: Strengths and Weaknesses of DOJ's Position

Strengths, from the standpoint of the argument DOJ is making:

• It states a clear, citable doctrinal position — that LLM training is "extraordinarily transformative" — in a case whose summary judgment ruling is likely to carry persuasive weight for other AI copyright suits.
• It frames the competitive consequences directly, arguing that licensing requirements would advantage only the largest technology companies, a point aimed at market structure rather than the parties' briefing alone.
• It adds a stated national security rationale to the fair-use debate, broadening the government's interest beyond copyright doctrine.

Weaknesses and open questions:

• The filing carries no binding legal authority; Judge Stein is free to reach a different conclusion regardless of DOJ's position.
• DOJ was not part of the discovery process, a gap Topic raised directly in disputing the filing's factual grounding.
• Summary judgment briefing was already due three days after the filing, leaving limited time for the parties to respond to DOJ's arguments before that deadline, so its influence on those particular filings is uncertain.
• Topic's framing — "If the administration's position was accepted, it would result in an unprecedented, uncompensated transfer of IP rights from news organizations to tech companies" — signals that publishers plan to contest the filing's premises rather than treat it as settled ground.

Outlook

DOJ's filing arrives alongside two related but distinct developments that point to a broader administration posture toward AI, without being part of this case. Anthropic reached a $1.5 billion copyright settlement in a separate matter before Judge William Alsup — a different case, a different judge, and not part of the consolidated OpenAI copyright litigation. In December 2025, President Trump signed an executive order aimed at establishing a federal AI standard applicable across all 50 states. Together with the Statement of Interest, these developments suggest an administration favoring fewer state-by-state constraints and a permissive federal fair-use posture for AI training, though each remains a separate legal action with its own scope and parties.

Whatever Judge Stein rules on the pending summary judgment motions, the decision is likely to draw attention from the growing set of AI training-data copyright suits working through other courts, now that DOJ's position is part of the public record. But the ruling will turn on the parties' own summary judgment briefing and the evidentiary record built during discovery — a record DOJ did not participate in creating.

Conclusion

DOJ's Statement of Interest is a notable public alignment between the federal government and OpenAI's fair-use defense, delivered just before summary judgment briefing in a closely watched copyright case. It is not a ruling, it does not bind Judge Stein, and it arrived without DOJ having taken part in discovery — a limitation opposing counsel has already raised publicly. Readers tracking AI copyright litigation, publishers negotiating licensing terms, and AI companies watching for regulatory signals should treat the filing as one data point in an unresolved case, best read alongside the summary judgment briefing due over the following weeks.

Editor's Verdict

DOJ Backs OpenAI's Fair-Use Case in NYT Copyright Suit is a workable proposition that fills a clear gap, even if it doesn't fundamentally change the landscape.

The strongest case for paying attention: states a clear, citable doctrinal position — LLM training as "extraordinarily transformative" — in a high-profile case likely to be referenced across other AI copyright suits. That alone raises the bar for what readers should expect in this space. Reinforcing that, frames the competitive consequences directly, arguing licensing mandates would advantage only the largest AI companies over publishers and smaller competitors alike — practical value rather than just headline appeal. The broader signal worth registering is straightforward: DOJ's Statement of Interest is not an amicus brief — it is the specific legal vehicle federal agencies use to state a position without becoming a party, despite several outlets calling it a "brief." On the other side of the ledger, one constraint is real rather than a marketing footnote: carries no binding legal authority; Judge Stein can reach a different conclusion regardless of DOJ's position. It should factor into any serious decision. Layered on top of that, DOJ was not part of the discovery process, a gap Matt Topic cited directly in disputing the filing's grounding — which narrows the set of teams for whom this is an obvious yes.

For AI industry watchers, strategy teams, and decision-makers tracking platform shifts, the smart move is to track its trajectory and revisit once the rough edges are filed down. For everyone else, the safer posture is to monitor coverage and revisit once the use cases that matter to your team are demonstrated in the wild.