Nvidia Sued by Authors Over Use of Books to Train AI Without Permission
• Nvidia sued by 3 authors - Brian Keene, Abdi Nazemian, and Stewart O'Nan - for using their copyrighted books without permission to train its NeMo AI platform
• Lawsuit alleges Nvidia used a dataset of 196,640 books to help train NeMo to simulate written language
• Seeking unspecified damages for authors whose works were used to train NeMo's large language models in past 3 years
• Part of growing litigation over AI technology's use of copyrighted content without permission
• Adds to lawsuits facing other AI companies like OpenAI (ChatGPT) and Microsoft over generative AI