Two Newsrooms Sue Microsoft and OpenAI While Still Taking Their Money
The Seattle Times files copyright infringement claims against funders of its own journalism programs, reflecting deepening tensions over AI training data.
When Your Benefactor Becomes Your Defendant
The Seattle Times filed a copyright infringement lawsuit this week against Microsoft and OpenAI, companies that have simultaneously funded the newspaper's journalism projects and fellowships. The legal action, filed jointly with New York-based Newsday, argues that generative AI systems trained on news content threaten to render the journalism industry "broken beyond repair."
At DailyTechWire, we've tracked the evolving dynamic between publishers and AI labs across North America and Europe for two years. This case stands out not for its legal arguments, which mirror earlier complaints, but for the financial entanglement it exposes. The paradox of a newsroom accepting grants from the same entities it accuses of existential harm illuminates the uncomfortable dependencies now woven through media economics.
The complaint describes AI products as "rapacious consumers, devouring human-authored content and delivering back to the world copies and derivative imitations." It invokes the metaphor of a snake eating its own tail, warning that generative systems could "destroy the very organizations" producing the training material they depend on.
The Expanding Legal Front
The Seattle Times and Newsday join a lengthening roster of American news organizations pursuing copyright claims against frontier AI labs. The New York Times initiated the template in late 2023 with a lawsuit targeting both OpenAI and Microsoft, its primary investor and infrastructure partner. That case remains in discovery, but it opened a path for smaller regional outlets to assert similar claims.
What distinguishes the current wave from earlier disputes over web scraping or search aggregation is the scale of reuse alleged. Publishers contend that large language models don't merely link to or excerpt their work but internalize vast corpuses of articles, enabling systems to generate text that competes directly with original reporting without compensation or attribution.
The complaint filed this week does not specify damages, but earlier publisher lawsuits have framed the harm in both economic and existential terms: lost subscription revenue, diminished traffic, and the long-term erosion of incentives to fund investigative work if AI systems can synthesize and repackage it instantly.
Microsoft's Measured Response
A Microsoft spokesperson characterized the company as "surprised by the lawsuit" and expressed willingness to "sit down and explore solutions to this type of dispute." The statement, relayed through regional technology press, stops short of acknowledging infringement but signals openness to negotiated licensing frameworks, a posture Microsoft has adopted in parallel conversations with book publishers and photo agencies.
OpenAI has not issued a public comment on the Seattle Times case. The lab has previously argued that training on publicly accessible web content constitutes fair use under U.S. copyright doctrine, a defense that will be tested as these cases advance through federal courts.
The dual role Microsoft plays complicates the optics. The company operates both as a technology platform provider, offering Azure infrastructure to OpenAI, and as a direct investor in news organizations through philanthropic arms and product partnerships. Those partnerships have included funding for data journalism initiatives, fellowships for early-career reporters, and cloud credits for digital transformation projects at regional outlets including The Seattle Times.
The Uncomfortable Economics of Patronage
The financial relationship between The Seattle Times and Microsoft predates the current AI boom. Microsoft has supported journalism programs in the Pacific Northwest for years, part of a broader corporate strategy to bolster civic institutions in its home region. Those grants have funded environmental reporting, data visualization tools, and paid internships.
That history now creates an awkward tableau: a newsroom accepting operational support from an entity it simultaneously accuses of undermining the sustainability of its business model. The lawsuit does not call for an end to those funding arrangements, and neither party has indicated whether the legal action will affect ongoing programs.
Industry observers note that this dynamic is not unique to Seattle. Across the United States and Europe, news organizations find themselves dependent on a narrow set of technology companies for advertising revenue, cloud infrastructure, reader analytics, and philanthropic grants. When those same companies become competitors or alleged infringers, the resulting legal and business decisions carry inherent contradictions.
What the Case Could Clarify
Legal experts expect the Seattle Times complaint to hinge on two questions that remain unsettled in U.S. copyright law. First, does the use of copyrighted text to train a generative model constitute reproduction or adaptation, acts that require a license? Second, if it does, can defendants invoke fair use based on the transformative nature of the output and the public benefit of AI systems?
Courts have not yet ruled definitively on either question in the context of large language models. The New York Times case, currently in the Southern District of New York, is likely to produce the first binding precedent, but appeals could extend the timeline for years.
In the interim, some publishers have opted for licensing deals rather than litigation. Axel Springer, News Corp, and the Associated Press have all signed agreements with OpenAI, trading access to archives in exchange for undisclosed fees and, in some cases, product integrations. The Seattle Times and Newsday have chosen the opposite path, betting that courts will recognize a violation substantial enough to warrant damages and injunctive relief.
The Tail-Eating Metaphor and Its Limits
The lawsuit's central metaphor, generative AI as a self-consuming serpent, captures a genuine concern: if models can produce plausible substitutes for original journalism at near-zero marginal cost, the economic foundation for newsgathering erodes. Readers who once visited a newspaper's website for a weather update, election result, or product review might instead ask an AI assistant, depriving the publisher of the ad impression or subscription nudge that funds the next round of reporting.
Yet the metaphor also glosses over complexity. Not all AI-generated text competes with the sources it was trained on. A model that answers a technical question by synthesizing explanations from a dozen articles may reduce traffic to any single publisher but also surfaces information that no one outlet covered comprehensively. The net effect on the journalism ecosystem depends on use cases, user behavior, and the design of the systems themselves, variables the lawsuit does not dissect in detail.
At DailyTechWire, we've observed that the most vulnerable publishers are those relying on high-volume, low-differentiation content: commodity news, basic explainers, and search-optimized articles. Investigative work, deep narrative features, and scoops remain harder for models to replicate, at least with current architectures. The question facing the industry is whether the revenue mix can shift fast enough to reward the irreplaceable while automating or abandoning the routine.
Regulatory and Market Pressure Building
The Seattle Times lawsuit arrives as policymakers in Washington, Brussels, and Beijing debate frameworks for AI training data. The European Union's AI Act includes transparency requirements for training datasets, and proposed legislation in California would mandate disclosure of copyrighted sources. Neither regime resolves the underlying copyright question, but both increase compliance costs and legal risk for labs that train on unlicensed material.
Market dynamics are also shifting. OpenAI's reported negotiations with publishers have grown more frequent as the company prepares for a potential public offering and seeks to preempt regulatory intervention. Microsoft, meanwhile, faces scrutiny from antitrust regulators over its investments in AI startups and its bundling of AI features into enterprise software, adding another layer of legal exposure.
The outcome of the Seattle Times case will influence how aggressively other regional outlets pursue similar claims. If plaintiffs secure a favorable ruling or settlement, expect a cascade of filings from mid-sized newspapers, magazine publishers, and trade press. If courts side with the defendants on fair use grounds, the window for extracting licensing fees may close, leaving publishers to compete with AI-generated substitutes on unequal terms.
The Paradox Unresolved
For now, The Seattle Times continues to operate with Microsoft funding while its attorneys argue in federal court that the company's AI partners are destroying the industry. Neither side has framed this as hypocrisy. The newsroom likely views the grants as partial restitution for harms already inflicted, while Microsoft may see the support as goodwill that insulates it from accusations of indifference to journalism's fate.
The contradiction, however uncomfortable, reflects the broader reality facing legacy media in the age of generative AI. Publishers need the revenue, tools, and distribution that technology platforms provide, even as those same platforms develop products that could hollow out their core business. Lawsuits offer one path to rebalancing that relationship, but they come with the risk of severing partnerships that many newsrooms can't yet afford to lose.
Whether the courts will force a reckoning, or whether market forces and backroom deals will determine the terms, remains an open question. What's clear is that the journalism industry's relationship with AI labs has moved from wary coexistence to open conflict, with the Seattle Times lawsuit serving as a particularly vivid case study in the contradictions that arise when survival and principle collide.
