Seattle Times and Newsday Sue OpenAI and Microsoft Over AI Training Content

The Seattle Times and Newsday filed a lawsuit on Friday against OpenAI and Microsoft, accusing both companies of systematically copying editorial articles to train artificial intelligence models without consent or compensation. It's the latest legal action targeting how AI companies acquired data to build their systems, following The New York Times' 2023 lawsuit and a June lawsuit from a coalition of nearly 400 local newspapers.

What the publishers allege

According to the lawsuit, OpenAI and Microsoft spent years scraping content from both newspapers' websites, including articles available only to paying subscribers, and incorporated it into datasets used to train and run products such as ChatGPT, Microsoft Copilot, and the AI features built into the Bing search engine. Lawyers representing the Seattle Times and Newsday argue that these systems can reproduce passages of their text and closely paraphrase entire articles, which reduces the number of readers visiting the outlets' websites and encourages them to cancel their subscriptions.

The lawsuit also includes a trademark infringement claim, alleging false attribution of content to the newsrooms whose material was used to generate chatbot responses. For publishers, this isn't just a copyright issue but also one of brand credibility, when AI purportedly quotes them while actually paraphrasing or getting facts wrong.

The numbers behind the lawsuit

A key economic argument in the lawsuit is the 47 percent year-over-year drop in search engine referral traffic to mid-sized publishers recorded in December 2025. The plaintiffs' lawyers say this is a direct consequence of AI-generated answers that satisfy users' curiosity without redirecting them to the source website.

The lawsuit's demands include the destruction of copies of the disputed works and all AI training datasets and models containing them, as well as monetary damages, the exact amount of which the parties have not yet disclosed.

We feel strongly that we must defend our content, which we spend millions of dollars a year to produce, from being used without our consent or compensation - Alan Fisco, President and CEO of the Seattle Times

OpenAI and Microsoft respond

An OpenAI spokesperson repeated the company's established defense, arguing that its models are trained on publicly available data and fall within the bounds of fair use, the US copyright exception for transformative use. Microsoft responded more cautiously, saying it remains open to discussions.

While we're surprised by the lawsuit, we appreciate the importance of local journalism and we're always happy to sit down and explore solutions - Microsoft spokesperson

The broader legal fight

The case is part of a growing wave of lawsuits by publishers against AI companies. The New York Times sued OpenAI and Microsoft over the same issue back in 2023, and in June 2026 a coalition of nearly 400 local US newspapers filed a similar suit, accusing the companies of systematically and covertly scraping websites and stripping copyright management information from the copied material.

The situation is complicated by the position of the US government, which in September 2026 officially sided with OpenAI in a similar dispute for the first time, arguing that training models on copyrighted text does not, as a general matter, constitute infringement. It's the first time the federal administration has taken a position in a case of this kind, which could affect how other proceedings unfold, including the one filed by the Seattle Times and Newsday.

For Polish publishers and content creators, the case carries precedent-setting weight even though it's playing out in the US legal system. The outcome of these proceedings could shape global standards for compensating training data, standards that in practice also underpin the models used in Europe, including in Poland.

Further dates in the case, including the defendants' first procedural responses, should be set in the coming weeks. Both sides have a strong interest in a swift resolution, since the outcome could influence dozens of other pending and planned lawsuits against AI companies brought by news publishers worldwide.