AI companies are anonymously buying and destroying millions of books through middleman services to avoid headlines about…
Major AI companies are reportedly using third-party services to anonymously purchase and destroy millions of physical books for training data, aiming to avoid negative public relations. This practice, highlighted by reporting on Anthropic's methods, involves sourcing rare and out-of-print books to create unique training datasets for large language models, with the destruction of books being deemed 'fair use' under copyright law.
This article was reported and published by pcgamer.com. feed.gg links to it as part of a story cluster — full text, images and rights remain with the publisher.