The Legal Fine Lines of Fair Use and Generative AI

Training sets for large language models (LLMs) is changing how we think about copyright but the law hasn’t quite caught up yet. Preserving the incentive for human beings to create artistic and scientific works is the lodestar of copyright law. Recent court decisions in many ongoing cases related to the use of copyrighted materials for AI training data continue to stir conversation and controversy around the four fair use factors. Based on these decisions, training on copyrighted texts is now endorsed as fair use by at least one court, which is a major gain for AI developers. However, other courts may disagree, and using pirated sources to train the AI models still carries risk. Ultimately, there is no distinct, final precedent stating all AI training is or is not fair use and facts are evolving at a rapid pace. As noted in the May 2025, U.S. Copyright Office publication Copyright and Artificial Intelligence Part 3: Generative AI Training, “These groundbreaking technologies should benefit both the innovators who design them and the creators whose content fuels them, as well as the general public.”

In this sharp, timely webinar, legal experts unpack the evolving landscape of fair use as it applies to generative AI: What counts as truly “transformative”? When does training an AI become copyright infringement? And how could new case law and regulation reshape what’s permissible? Whether you’re in publishing, law, library science, or tech, this conversation offers a guide to the legal frontiers we're beginning to explore.