A group of US local news organizations led by Emmerich Newspapers has sued Microsoft and OpenAI, alleging the companies used tens of thousands of copyrighted news articles to train large models without authorization or compensation. The complaint says copyright management information such as author bylines, publisher names and copyright notices was stripped from the articles before training, and that their content has been reproduced by AI systems in near-verbatim form. The plaintiffs cite the US Copyright Act and the Digital Millennium Copyright Act, seeking damages and an injunction requiring the removal of their works from training data.