Authors Guild Lawsuit Challenges OpenAI Over Copyrighted Book Training Data
A landmark copyright lawsuit brought by the Authors Guild against OpenAI and Microsoft questions the legality of using copyrighted books to train large language models. The case explores how traditional copyright law applies to modern generative AI systems.
The Authors Guild and several individual authors sue OpenAI in a landmark copyright case that centers on the unauthorized use of copyrighted books to train large language models. The plaintiffs argue that OpenAI copies massive amounts of protected literary works without permission to develop and commercialize its GPT models. They later amend their complaint to include Microsoft, claiming the tech giant plays a significant role in deploying and profiting from these AI systems.
This litigation raises fundamental questions about how traditional copyright law applies to the modern era of generative artificial intelligence. The legal theory extends beyond the initial act of training, as the authors also point to allegedly infringing AI outputs and the broader economic harm that these systems cause to writers and book-licensing markets. The case addresses whether the ingestion of text for machine learning constitutes fair use or an illegal reproduction of intellectual property.
The dispute evolves into a broader multidistrict litigation (MDL) in New York, consolidating various copyright claims against OpenAI and Microsoft. As the MDL remains active, the court allows a direct copyright claim based on AI-generated outputs to move forward, marking a significant milestone in the legal battle. Ongoing discovery disputes keep the process moving as both sides prepare for further arguments over the boundaries of AI and copyright law.