Critics allege AI firms knew copyright risks per leaked docs

Legal critics allege that leaked internal communications demonstrate AI companies understood copyright law yet proceeded with unauthorized training data usage. Matthew Dessem argued on Bluesky that such documents contradict industry claims of legal ignorance, suggesting instead a calculated disregard for intellectual property rights. These allegations emerge amid ongoing federal lawsuits where plaintiffs seek to prove willful infringement to secure statutory damages. AI defendants have consistently maintained their training practices constitute fair use and deny intentional wrongdoing. The dispute centers on whether internal knowledge of legal risk negates good faith defenses in court. If verified, these communications could significantly impact judicial assessments of fair use factors. Legal experts note that proving willfulness remains a high evidentiary bar requiring authenticated documentation. The controversy highlights intensifying scrutiny over how AI firms document and manage copyright compliance during model development.