The U.S. government has sided with OpenAI in a legal case brought by the New York Times over OpenAI training artificial intelligence (AI) models on paywalled New York Times articles.
“The United States has a strong interest in this Court rejecting any argument that training LLMs on copyrighted texts violates copyright law,” government lawyers said in a Sept. 1 brief filed with federal court in New York City. LLMs, or large language models, are trained on documents and images.
The reasons behind the government’s position, the lawyers said, include ensuring AI systems are capable of assisting the military and helping “level the playing field between mainstream and independent publishers” by enabling “authors with limited resources” to compete by, for example, using the models to generate images to accompany articles….