Creative Ownership in the Age of AI

2026-02-12

AI Summary

Current copyright law struggles to determine whether generative AI infringes rights when it mimics artistic style without directly copying content. This work proposes a novel infringement criterion based on dependence on training data: generation is deemed infringing if the output could not have been produced without a specific work being present in the training set. To formalize this notion, the authors model the generative process as a closure operator for the first time, capturing the structural properties of legally permissible generation. Integrating probabilistic tail analysis, they further demonstrate how the tail behavior of the underlying creative distribution governs regulatory efficacy. Their analysis shows that under light-tailed distributions, the influence of any individual work vanishes as model scale increases, rendering regulation ineffective; conversely, under heavy-tailed distributions, regulation can sustainably constrain generative behavior over time.

Technology Category

• Natural Language Processing: Generation
• Philosophy and Ethics of AI: AI & Law, Justice, Regulation & Governance
• Computer Vision: Generative Adversarial Networks (GANs) for Vision

Application Category

• Economics, Online Markets and Human Computation: Economic ramifications for generative AI infrastructure and applications
• Social Networks and Social Media: Generative AI / large language models and their impact on social systems
• Web Mining and Content Analysis: Web data quality in the era of algorithmically-generated content

Abstract

Copyright law focuses on whether a new work is"substantially similar"to an existing one, but generative AI can closely imitate style without copying content, a capability now central to ongoing litigation. We argue that existing definitions of infringement are ill-suited to this setting and propose a new criterion: a generative AI output infringes on an existing work if it could not have been generated without that work in its training corpus. To operationalize this definition, we model generative systems as closure operators mapping a corpus of existing works to an output of new works. AI generated outputs are \emph{permissible} if they do not infringe on any existing work according to our criterion. Our results characterize structural properties of permissible generation and reveal a sharp asymptotic dichotomy: when the process of organic creations is light-tailed, dependence on individual works eventually vanishes, so that regulation imposes no limits on AI generation; with heavy-tailed creations, regulation can be persistently constraining.

Problem

• copyright infringement
• generative AI
• creative ownership
• training corpus
• style imitation

Innovation

• generative AI
• copyright infringement
• closure operator
• permissible generation
• heavy-tailed distribution