Microsoft and OpenAI Face New Scrutiny Over AI Training Data

A coalition of prominent authors, including Nicholas Basbanes and Nicholas Gage, has launched a federal lawsuit against Microsoft and OpenAI, alleging that the companies utilized their copyrighted works to train generative AI models without consent or compensation, marking a significant escalation in the ongoing legal battle over artificial intelligence copyright.

The plaintiffs filed the class-action complaint in a Manhattan federal court, asserting that the technology giants systematically harvested millions of books to refine their large language models. The authors argue that the unauthorized use of their intellectual property constitutes a direct infringement of copyright law, threatening the economic viability of professional writing. By feeding protected works into algorithms that can mimic human style and generate derivative content, the companies have effectively built a commercial engine on the backs of creators who never authorized such use.

Legal representatives for the authors emphasize that this practice undermines the incentive structure of creative industries. While Microsoft and OpenAI have previously maintained that training AI models on public internet data qualifies as fair use, the plaintiffs contend that the deliberate extraction of specific, high-quality literary works transcends legal protections. This litigation seeks to establish a precedent regarding the commercial exploitation of human expression in the era of automated intelligence, potentially reshaping how technology firms approach data sourcing for future model development.