Training AI on copyrighted books sits in a legal gray zone that courts have yet to resolve. The core question is whether scraping published works for training data counts as fair use or infringement. For AI developers, the outcome could reshape how datasets are built and what consent is required.
Opening Kapyn…