Meta is facing allegations of using pirated content to train its Llama AI model. A 2023 lawsuit, Kadrey et al. v. Meta Platforms, claims the company used copyrighted material without authorization, including books and articles from piracy hubs like LibGen. Court documents reveal internal conversations suggesting Meta employees and CEO Mark Zuckerberg were aware of the practice.
Meta defends its actions under the “fair use” doctrine, arguing it used public materials for statistical modeling. This isn’t the first AI controversy – Apple faced similar scrutiny over its OpenELM model, though it clarified its AI training relies on licensed and public data. Major publishers like The New York Times have opted out of sharing content for AI training.
Source: 9to5Mac
