Seattle Times, Newsday sue OpenAI, Microsoft

The Seattle Times Co. and Newsday have filed a copyright infringement lawsuit against OpenAI and Microsoft (MSFT) in U.S. federal court in the Southern District of New York. The complaint, lodged on September 4, 2026, centers on the use of the newspapers’ journalism in training and operating a range of artificial intelligence products.

The plaintiffs allege that the companies scraped hundreds of thousands of articles from their websites, including material that sits behind digital paywalls. They claim these works were incorporated into datasets used to build and run systems such as ChatGPT, Microsoft (MSFT) Copilot and AI features in Bing.

According to the complaint, the AI models can reproduce passages from Seattle Times and Newsday articles or closely paraphrase their reporting. The publishers say the filing includes examples in which AI outputs contain near-verbatim excerpts of their work.

The newspapers argue that these capabilities may reduce the need for users to visit their websites or purchase subscriptions, because the AI products can supply answers that substitute for reading the original articles. They present this as harm flowing from the alleged unauthorized use of their content in training.

Seattle Times Co. and Newsday are seeking unspecified monetary damages from OpenAI and Microsoft. Beyond financial compensation, they are also asking the court to order the impoundment and possible destruction of copies of their works used in AI training.

The requested relief extends to any training datasets and AI models that incorporate the newspapers’ journalism. If granted, such orders could require the removal or destruction of data and systems that include their content.

OpenAI states that its models are trained on publicly available data and that its approach is grounded in the legal doctrine of fair use. This position reflects its broader stance that using large-scale datasets for training can be lawful when done under existing copyright frameworks.

Microsoft has said it was surprised by the lawsuit and indicated it is willing to discuss potential solutions with the publishers. Both companies’ responses frame the dispute as one that touches on how copyright law applies to modern AI development and deployment.