The Seattle Times Co. sued OpenAI and Microsoft on Friday in the Southern District of New York, joined by Newsday, accusing the companies of using the papers' work to train large language models. [1] The filing posted at 5:02 p.m. Pacific, after Friday's edition had closed. Saturday's wires caught up. MSM will sell another publisher copyright suit. The hometown paper is suing two of its own grantors the week the Justice Department backed fair use. A filing is not an injunction. Do not write that Microsoft cut the philanthropy. Do not write that the models will be destroyed.
President and CEO Alan Fisco told staff it "was not an easy decision." [1] "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." [1] The complaint cites industry data that search referral traffic to midsize publishers fell 47 percent year over year in December 2025. [1] It also accuses OpenAI of diluting trademarks by generating content falsely attributed to the Times and Newsday. [1] Harm claimed is not harm proved. Forty-seven percent is a cited industry figure, not a Seattle Times circulation audit.
A Microsoft spokesperson said the company was "surprised by the lawsuit," appreciated "the importance of the Seattle Times to our region," and was "always happy to sit down and explore solutions." [1] Reuters heard "the importance of local journalism" in the same sit-down line. [2] Print both. OpenAI said its models are trained on publicly available data and "grounded in fair use," which it said helps people and "delivers benefits such as empowering human creativity, science, and medical research." [1] Fair use is the defense. It is not a ruling. Reuters wrote that the complaint alleges scraping, including paywalled copy, into datasets used for ChatGPT, Copilot, and Bing's AI features, and that the products can reproduce passages or reduce the need to visit the sites. [2] Alleged is not found.
Tuesday, the Justice Department filed a brief supporting OpenAI in the New York Times case, saying AI training generally makes fair use of copyrighted material. [3] The brief appears to be the first time the U.S. government has weighed in on the wave of training suits. [3] Associate Attorney General Stanley Woodward said "AI dominance is critical" and that the administration would not accept "a plainly incorrect understanding of copyright law." [3] The government called training "extraordinarily" transformative. [3] The Times called the brief a siding with "trillion-dollar AI companies" at the expense of creators. [3] Same week, same Manhattan court. This suit is not that case. Do not recap Friday's wiki dump as this complaint.
Microsoft Philanthropies underwrites some Seattle Times journalism projects, the paper wrote, while maintaining editorial independence. [1] The Times, with other publishers, received a Lenfest Institute grant to hire a two-year AI fellow in 2024. Microsoft and OpenAI funded that fellowship with $10 million. [1] Both newsrooms were in the program. The Times has used the grant for some advertising functions and holds ChatGPT licenses for data analysis. [1] Grantor and defendant are the same names. Do not write the philanthropy was cut.
Microsoft has invested billions in OpenAI since 2019 and remains the primary cloud partner. [1] The companies worked together on the models behind ChatGPT and Copilot. [1] Microsoft said in April it is still a major shareholder. [1] Two years ago Alden-owned dailies sued OpenAI. [1] Last year 35 publishers representing more than 400 community papers sued as well. [1] Adjacent dockets are not this caption. The Times has a public AI newsroom policy: limited, labeled uses after risk review, no AI-generated stories. [1] Policy is not a license grant to the defendants.
The papers seek damages and destruction of training sets that used their work. [1][2] Reuters wrote that they also seek destruction of AI models incorporating the works. [2] Sought is not ordered. Anthropic's $1.5 billion authors' settlement is last year's adjacent docket, not this complaint. [1] Fisco said the suit was "not about impeding AI innovation" but about innovation that does not come "at the expense of our business model." [1] A business-model sentence is not an injunction. The hometown paper sued the grantors. The week DOJ backed fair use is the texture. The wiki file is a different Saturday.
-- DAVID CHEN, Beijing