In a pivotal development for the AI industry, a federal judge in California certified a class action lawsuit against Anthropic, an AI company, due to allegations of large-scale copyright infringement. The lawsuit involves the unauthorized use of millions of copyrighted books to train Anthropic’s Claude AI platform, a case that could potentially rival some of the largest copyright disputes in American legal history. JURIST reports that Senior District Judge William Alsup of the Northern District of California approved the class certification for authors whose works were obtained from pirate libraries such as LibGen and PiLiMi.
This decision comes after a recent mixed ruling on the application of fair use in AI training. While Judge Alsup had previously acknowledged that the transformation of copyrighted materials for AI training could be legally justified as “quintessentially transformative” fair use, he distinguished this from the illicit downloading of books from internet pirate sites. The court’s decision to certify the class focuses on authors whose copyrighted books were included in the millions of works Anthropic allegedly downloaded to develop its AI capabilities.
Under U.S. copyright law, the repercussions for willful infringement are severe, with penalties reaching up to $150,000 per work. Given the scale of the alleged infringement, which involves nearly seven million books, Anthropic faces potentially massive liabilities. The argument that the size of potential damages should prevent class certification was dismissed by the court, referencing longstanding precedent that upholds the purpose of class actions even when potential damages are considerable.
This case draws parallels with past high-stakes copyright litigation, such as the Napster case, where unauthorized downloading of copyrighted music led to substantial legal consequences. However, the sheer volume of works involved in the Anthropic suit sets it apart, with potential damages exceeding historical cases like Oracle’s $8.8 billion copyright claim against Google or Viacom’s billion-dollar suit against YouTube.
The implications of this lawsuit extend beyond Anthropic, potentially affecting the broader AI industry. Many AI companies employ similar methodologies for training their models, raising questions about the acquisition and use of copyrighted materials. The Verge highlights that a significant judgment could lead to an industry-wide reassessment of how AI models are trained.
Anthropic, which reportedly generates billions in revenue from its AI platform, now faces a legal landscape that could redefine data usage practices within the technology sector. The outcome of this case has the potential to establish key legal precedents and reshape the dynamics between copyright holders and AI developers.