DOJ Backs OpenAI in Landmark Copyright Dispute With The Intercept

Key facts

On September 2, 2026, the Justice Department entered the legal battle between OpenAI and digital publishers by filing a formal submission in federal court supporting the AI developer's position, as reported by Noah Hurowitz. The legal action directly addresses claims made by The Intercept and affiliated digital media platforms, which accused OpenAI of systematically scraping, copying, and processing thousands of copyrighted news articles to build and refine its generative AI systems.

In their initial filings, the news organizations alleged that OpenAI derived substantial commercial value from their original reporting while bypassing traditional licensing agreements. The media plaintiffs contended that using proprietary news copy to train competitive AI systems devalues published work, displaces web traffic, and threatens the economic viability of investigative journalism.

The Justice Department's brief counters those arguments by asking the court to find in favor of OpenAI. The government argued that the automated ingestion of public text for machine learning serves a transformative purpose rather than simple content duplication. According to the filing, imposing copyright liability on AI developers for ingesting digital text would impede technological innovation and create legal friction for the nation's broader technology infrastructure. The administration urged the presiding judge to interpret existing federal copyright law in a manner that allows AI developers to continue training advanced software on publicly accessible digital materials.

Why it matters

The Justice Department's intervention marks a critical juncture in the conflict between content creators and the artificial intelligence industry. By taking a side in litigation between private parties, the executive branch is seeking to influence judicial interpretation of federal copyright standards at a time when technology firms and media companies are locked in negotiations over the financial terms of the digital economy.

For news organizations, the government's position poses a direct threat to emerging licensing models. Over recent years, digital publishers have relied heavily on subscription revenues, direct syndication, and content licensing to fund newsrooms. If federal courts adopt the Justice Department's position and declare AI model training exempt from copyright infringement claims, media companies will lose critical legal leverage to negotiate commercial licensing deals with technology platforms. Without licensing revenue or court enforcement, publishers argue that AI systems will continue to absorb expensive, human-generated journalism to produce automated answers that directly substitute for original news sources.

Conversely, for artificial intelligence companies, executive backing provides major legal momentum. Training modern large language models requires petabytes of textual data, ranging from digitized books and encyclopedias to daily news reporting and internet forums. If courts were to require explicit licensing or monetary damages for every piece of copyrighted text used in training pipelines, AI companies could face legal liabilities reaching billions of dollars. The Justice Department’s brief underscores an explicit industrial strategy that prioritizes national technological dominance and rapid AI development over strict enforcement of traditional intellectual property claims by legacy publishers.

The dispute rests upon the interpretation of Section 107 of the U.S. Copyright Act of 1976, which governs the doctrine of fair use. Under federal law, courts evaluate four statutory factors when determining whether unauthorized use of copyrighted material is permissible: the purpose and character of the use, the nature of the copyrighted work, the amount and substantiality of the portion used, and the effect of the use upon the potential market for or value of the original work.

For decades, digital fair use jurisprudence was shaped by benchmark cases such as the 2015 ruling in Authors Guild v. Google. In that case, the U.S. Court of Appeals for the Second Circuit determined that Google's unauthorized scanning and indexing of millions of published books to create a searchable snippet database constituted transformative fair use. Google’s system provided public information about books without providing full-text substitutes that would harm book sales.

AI developers like OpenAI argue that large language models function similarly to Google Books or human readers: the models analyze statistical patterns, grammar, and facts across vast text libraries to learn human language rules, rather than serving as digital file repositories. Publishers, however, contend that generative AI represents a fundamentally different enterprise. Modern LLMs do not merely display snippets or search indexes; they generate complete text summaries, synthetic articles, and conversational outputs that can directly replace the need to visit original news websites, thereby harming the primary market for original journalism.

This legal confrontation expanded significantly between 2023 and 2026, as numerous copyright actions were filed across U.S. district courts. Outlets ranging from traditional newspapers to independent digital platforms filed suit against major AI labs, alleging that training LLMs on their archives without permission violated both copyright law and digital management provisions. While some legacy media conglomerates chose to negotiate direct, multi-year licensing deals worth tens of millions of dollars with firms like OpenAI, others turned to federal courts to establish binding legal precedent.

Reaction

The Justice Department's filing drew immediate reactions from press freedom advocates, legal scholars, and technology representatives. Media industry groups expressed severe disappointment, arguing that the executive branch is actively choosing to protect multi-billion-dollar technology conglomerates at the expense of independent journalism and content creators. Advocates for news organizations warned that granting tech platforms free access to news archives starves journalism of essential revenues required to maintain investigative reporting.

Conversely, technology trade organizations and artificial intelligence researchers welcomed the government's intervention. Industry groups maintained that machine learning models require comprehensive access to public internet data to operate effectively, and that treating data analysis as copyright infringement would stall scientific and commercial progress across the U.S. software sector.

Legal experts noted that while an amicus filing from the Department of Justice carries substantial persuasive weight in federal courts, federal judges operate independently and are bound strictly by statutory text and legal precedent rather than executive branch policy preferences. Observers expect the presiding judge to scrutinize the filing carefully during upcoming oral arguments.

What we don't know yet

Several fundamental elements of the case remain unresolved. It is not yet clear whether the presiding federal district court judge will accept the Justice Department's characterization of AI model training as fair use, or whether the court will allow the lawsuit to proceed through full discovery and trial.

Furthermore, technical transparency remains a significant point of contention. Because AI firms treat their training datasets and algorithmic architectures as highly guarded trade secrets, the exact quantity of proprietary articles from The Intercept and other co-plaintiffs ingested by OpenAI's specific model iterations—such as GPT-4—has not been fully disclosed in public court filings.

Additionally, it remains unknown whether other federal entities, such as the U.S. Copyright Office or the Federal Trade Commission, will issue statutory interpretations or regulatory guidelines that align with or contradict the stance taken by the Justice Department.

What to watch

In the coming months, court dockets will reveal critical procedural milestones in The Intercept v. OpenAI. Observers will monitor whether the federal district court rules on OpenAI's motion to dismiss or sets a schedule for summary judgment motions where the DOJ's brief will be formally debated.

Simultaneously, legal observers are watching parallel copyright cases, including high-profile suits brought by national newspapers and book authors in federal courts across New York and California. Rulings in those parallel proceedings could create conflicting circuit court precedents, potentially setting up a final resolution before the U.S. Supreme Court.

Finally, stakeholders will watch Capitol Hill for any legislative responses. While federal courts deliberate on existing statutes, members of Congress have debated proposals to establish statutory licensing systems, AI transparency mandates, or media compensation frameworks that could override judicial decisions regarding AI training data.

This report is based on original reporting by Noah Hurowitz.