Seattle Times, Newsday Sue OpenAI, Microsoft

Lawsuit Filed

The Seattle Times and Newsday have jointly filed a lawsuit against OpenAI and Microsoft, alleging copyright infringement. The two news organizations claim the companies used their published articles without permission to train AI models, including ChatGPT and Microsoft’s Copilot.

The lawsuit, filed in a U.S. federal court, seeks to hold the tech firms accountable for what the publishers describe as unauthorized copying of their protected journalistic work. According to the complaint, the defendants scraped and reproduced content from the newspapers’ websites to build and improve their generative AI systems.

Key points in the filing include:

• Direct infringement of copyrighted articles.
• Unlawful reproduction of headlines, excerpts, and full stories.
• Failure to obtain licenses or pay compensation.

The publishers argue this use undermines their business models and the value of original reporting. They are seeking damages and a court order to halt further use of their content in AI training. The case adds to growing legal pressure on AI developers over the unlicensed use of news media.

Demand for Destruction

Beyond monetary compensation, the lawsuit seeks an unprecedented remedy: the destruction of the AI models themselves. The newspapers argue that because their copyrighted works were integral to the training process, the resulting neural networks are irreparably tainted. They contend that simply paying a licensing fee would not undo the harm, as the models’ very architecture and parameters were shaped by the unauthorized use of their content.

This demand extends to the underlying training datasets and any derivative systems built upon those models. The legal theory posits that the models function as unauthorized derivative works, making their continued operation a daily infringement. If granted, such an order would force the tech companies to delete or retrain systems that have cost hundreds of millions of dollars to develop.

While critics view this as an extreme and impractical measure, the plaintiffs frame it as the only true deterrent. As the case progresses, this request stands as a stark threat to the foundational economics of generative AI development.

Core Infringement Claim

At the heart of the lawsuit is the allegation that OpenAI and Microsoft used the newspapers’ articles without permission to train their AI models. The complaint asserts that the companies copied millions of copyrighted news stories to build and improve systems like ChatGPT and Bing Chat. According to the publishers, this unauthorized use directly violates copyright law, as the articles were not licensed or compensated for in any way.

The plaintiffs argue that the AI models rely on this content to generate responses, often reproducing or paraphrasing the original reporting without attribution. The claim focuses on the fact that the newspapers’ journalism is valuable, original work, and that using it to train commercial AI systems constitutes infringement. The lawsuit seeks to hold the tech companies accountable for what the publishers describe as systematic copying on a massive scale, rather than incidental or fair use of the material.

Potential Implications

This case could set a significant precedent for how AI companies use copyrighted news content in their training data. If the court rules in favor of the plaintiffs, it may compel AI developers to secure licenses or pay for using news articles, altering the economics of model development.

More broadly, a ruling against the AI firm could establish that scraping and reproducing copyrighted material without permission is not “fair use,” even when used for machine learning. This would affect not only news publishers but also authors, photographers, and other content creators whose work appears online.

Conversely, a decision for the AI company might embolden others to train on copyrighted data without compensation, potentially undermining the business models of news organizations. The outcome may also influence ongoing negotiations between publishers and tech firms, as well as future legislation. Legal experts will watch closely to see if the court addresses the distinction between transformative use and direct competition with the original source material.