The Seattle Times and Newsday have sued OpenAI and Microsoft, alleging that their journalism was copied into AI training data without permission and is sometimes reproduced in chatbot answers. Microsoft was named because its Copilot products use OpenAI technology.
The publishers are asking for more than financial damages. Their requested remedies include destroying copies of their articles, the training datasets containing them and AI models that incorporate the material. A court would need to decide both whether infringement occurred and whether such a broad technical remedy is appropriate.
The case joins lawsuits from The New York Times, Ziff Davis, Merriam-Webster and Encyclopedia Britannica, as well as litigation involving nearly 400 local newspapers. Publishers argue that answers generated from their reporting can reduce visits and subscription revenue, while AI companies have generally defended model training under copyright doctrines such as fair use.
Neither OpenAI nor Microsoft immediately commented to The Verge. The complaint therefore adds another claim to an unsettled legal fight rather than resolving how training, retrieval and generated excerpts should be treated. The demand to delete affected models raises a particularly consequential question because removing specific training material after a model is built is technically and operationally difficult.