Regional Newspapers Sue OpenAI and Microsoft Over AI Training

Two regional U.S. newspapers have joined the growing legal fight over AI companies' use of copyrighted journalism, filing suit against OpenAI and Microsoft in federal court in Manhattan.

The Seattle Times and Long Island-based Newsday allege that their reporting was incorporated into commercial AI systems without authorization or payment, according to Reuters, which first reported the lawsuit on September 5. The complaint targets technology behind OpenAI's ChatGPT as well as Microsoft's Copilot and Bing products.

Beyond training data

The publishers' claims extend beyond the initial training of AI models. They argue that AI products can now deliver information derived from their articles directly to users, potentially eliminating the need for readers to visit the newspapers' own websites. This represents a fundamental shift from how search engines traditionally operated—by directing traffic to publishers rather than replacing them.

The lawsuit alleges that the disputed material includes journalism protected behind subscription paywalls, raising questions about whether AI developers can freely use content that publishers have explicitly chosen to restrict.

Why it matters

This case signals that copyright disputes over AI training have moved beyond elite national publications into the regional news sector, where profit margins are thinner and the economic stakes potentially more severe. If courts side with publishers and order destruction of AI models or datasets containing copyrighted material—as The Seattle Times and Newsday are requesting—the implications could reshape how AI companies access training data across the industry.

The economics of journalism versus AI

Seattle Times CEO Alan Fisco told employees that the newspaper invests millions of dollars annually in original reporting and must defend that investment against uncompensated use, Reuters reported. The case highlights a collision between two business models: news organizations that spend heavily to gather information, and AI developers that require vast amounts of text to build useful products.

Publishers historically relied on intermediaries like search engines to drive readers to their sites, supporting subscription and advertising revenue. Generative AI threatens that model by answering questions within the technology company's interface, potentially capturing value that would otherwise flow to the original source.

Company responses

OpenAI disputes the legal foundation of such claims. A company spokesperson told Reuters that its models are built using publicly available information under fair-use principles, though the company did not address the specific allegations from the two newspapers.

Microsoft struck a more conciliatory tone, expressing surprise at the lawsuit while acknowledging the importance of local journalism and indicating openness to discussing resolution, according to Reuters.

Broader legal landscape

The Seattle Times and Newsday are entering a crowded field of AI copyright litigation. The New York Times sued OpenAI and Microsoft in 2023 over similar allegations, a case that remains pending. Other copyright owners have pursued claims against AI developers including Anthropic and Meta Platforms.

The expansion of these disputes into regional publishing is particularly significant because local newspapers operate with fewer resources than national outlets, making the potential economic impact of AI-generated content more acute.

These details were first reported by Reuters.