OpenAI Gains U.S. Backing in New York Times Copyright Fight
The AMW Read
The Justice Department's intervention materially strengthens the fair-use policy case for a top model lab, but it neither binds the court nor resolves the underlying training-data litigation.
OpenAI Gains U.S. Backing in New York Times Copyright Fight
The Trump administration filed a court letter supporting OpenAI’s fair-use defense in The New York Times’ copyright lawsuit. The Times sued OpenAI and Microsoft in 2023, alleging their models were trained on its articles without permission. The government argued that training large language models on copyrighted works is highly transformative, that OpenAI’s systems do not materially compete with Times articles, and that a restrictive ruling could weaken U.S. AI leadership. The letter is not binding on Judge Sidney H. Stein, but legal experts told WIRED it is likely to carry weight.
The intervention raises the stakes beyond one publisher dispute. The central market question is whether copyrighted text remains a licensable upstream asset for model training or can generally be treated as fair-use input for systems that transform it into new outputs. The administration’s position supports the latter view, while the Anthropic litigation shows that a fair-use finding for training may still coexist with liability for acquiring source material through piracy. That distinction leaves data provenance, rather than only model behavior, at the center of the legal risk.
For builders, the immediate implication is to document how training and retrieval data were obtained, licensed, and separated from potentially infringing copies. For investors, a favorable outcome for OpenAI would reduce one category of legal overhang for general-purpose model developers, but it would not settle the wider publisher, author, and music-rights cases now testing the limits of fair use.
