Fair Use In AI Training: Divergent Approaches In Recent Court Decisions
Fair Use In AI Training: Divergent Approaches In Recent Court Decisions
Introduction
The rise of generative AI has led to a surge in copyright infringement cases, with courts grappling to apply traditional fair use principles to novel technologies. Two recent decisions, Bartz v. Anthropic PBC and Kadrey v. Meta Platforms, highlight the differing perspectives on fair use in AI training.
Factual Background
The aforementioned cases involved author-plaintiffs whose books were copied to train large language models (LLMs) without permission. Anthropic's 'Claude' and Meta's 'Llama' were trained on unauthorized digital copies from 'shadow libraries'.
Procedural Background
The courts applied the four-factor fair use test under 17 U.S.C. § 107, examining the purpose and character of the use, the nature of the copyrighted work, the amount and substantiality of the work used, and the effect on the market for the work.
Reasoning & Analysis
The judges found that training on lawfully acquired materials is transformative and acknowledged the creative nature of novels weighs against fair use. However, they diverged on the treatment of pirated books and market harm. Alsup distinguished between pirated and legitimately acquired books, while Chhabria viewed all copies as part of a transformative process.
Divergence on Market Harm
The courts differed significantly on market harm, with Alsup dismissing concerns about AI-generated substitutes and Chhabria warning that future plaintiffs could succeed by showing market harm. The US Copyright Office expressed concerns similar to Chhabria's, noting that AI-generated content can dilute the market due to volume and speed.
Implications
The contrasting approaches in Bartz and Kadrey highlight the fluidity of legal boundaries in AI training. A potential Ninth Circuit ruling could shape the future of generative AI, and courts will soon face cases that force a definitive ruling on whether AI-induced market dilution defeats a fair use defense.
Final Outcome
Both courts ultimately found fair use, but their reasoning diverged. Anthropic has filed a motion for an interlocutory appeal, which could lead to a federal appeals court review and potentially shape the future of AI-related copyright law.
