The rise of generative AI technologies has introduced unprecedented challenges to copyright law, particularly around the fair use of copyrighted works in AI training processes. Generative AI tools, such as ChatGPT, are trained on vast datasets that often include copyrighted material, typically without the consent of authors or compensation for use. This widespread, unauthorized use has led to legal disputes, with plaintiffs asserting that using protected texts in training AI models constitutes copyright infringement. This Note examines the application of the fair use doctrine to generative AI, analyzing each of the four statutory factors to demonstrate that generative AI’s commercial replication of copyrighted content is not transformative, harms the market for original works, and should not qualify as fair use. To address these issues, this Note proposes a blanket licensing scheme as a policy solution to balance the interests of copyright holders and AI companies. Such a scheme would ensure compensation for authors while legally permitting AI companies to access necessary training data, and therefore foster a sustainable partnership between creators and the AI industry.
