Show summary Hide summary
The Trump administration has thrown its weight behind OpenAI in the copyright dispute with the New York Times, filing a federal brief that frames the case as a test of how the law should treat the training of artificial intelligence systems. The government’s intervention raises the stakes for publishers, tech companies and courts weighing how existing copyright rules apply to the development of large language models.
A federal voice in a high‑profile fight
In a court filing made public this week, the United States government sided with OpenAI against the New York Times, which has alleged that AI firms copied news articles to train chatbots without permission. The filing tells the court the outcome could reshape the legal landscape for AI research and commercial products.
Boxing champions ditch long runs for high-intensity training: new era in fight prep
Czech nationals among victims of Pinedale-area crash: two confirmed dead
The brief does not decide the case for the judges, but it signals how federal authorities view core legal questions: whether ingesting copyrighted works for model training qualifies as infringing or falls within recognized protections like fair use. That interpretation could affect dozens of lawsuits now pending and future claims by content owners.
Why this matters now
The government’s position comes as generative AI has become embedded in search, customer support, news summaries and creative tools. A legal ruling that treats training as actionable copying would force model builders to negotiate licenses or curb how they collect data — potentially slowing product development and increasing costs.
At the same time, a ruling that favors AI developers would unsettle publishers who rely on licensing and subscriptions for revenue. For news organizations already navigating shrinking ad and subscription margins, a decisive court ruling could influence their bargaining power and strategies for digital distribution.
- For AI firms: A favorable precedent would reduce legal uncertainty and could lower the cost of data acquisition and model training.
- For publishers: An adverse result could weaken the ability to demand licensing fees or to control use of their archives.
- For courts and regulators: The ruling will likely guide how copyright law is applied to new technologies and may prompt legislative attention.
Positions and pushback
OpenAI and other technology companies have argued that using texts to teach models is a necessary technical step and that the resulting systems produce transformative outputs rather than reproductions of copyrighted works. Publishers counter that automated scraping of articles, followed by the generation of text that can closely mimic reporting, undercuts the market for original journalism.
Legal scholars differ on how clear the law is. Some say existing doctrines like fair use can accommodate AI training; others contend courts must develop new rules or Congress should act to provide clarity. The government’s brief, while influential, is one voice among many in a complex legal debate.
Possible outcomes and next steps
What happens next depends on the court’s reading of both statutory law and precedent. The case could resolve at a district-court level, but given its broader implications, an appeal to an appellate court or even the Supreme Court is conceivable.
Whatever the judicial outcome, the dispute will shape industry behavior: companies may pursue voluntary licensing deals, change data‑handling practices, or push for clearer legislative standards. Publishers are likely to press for stronger protections and potentially new revenue models if they feel the law does not adequately safeguard their content.
Until judges issue a ruling, the filing by the Trump administration adds a significant legal and political dimension to a dispute that goes to the heart of how copyright will coexist with rapidly advancing AI technology.
