Rethinking Fair Use in the Age of AIGC: Balancing Copyright Protection and Technological Innovation

Abstract

The rapid advancement of generative artificial intelligence (AI) is fundamentally reshaping the paradigm of content creation, yet the processes of data input, model training, and content generation entail significant risks of copyright infringement. Currently, Article 24 of China's Copyright Law, with its strict restrictions on eligible subjects and permissible purposes, proves increasingly ill-suited to the complex realities of artificial intelligence generated content (AIGC) creation. This article identifies three fundamental dilemmas within the existing legal framework: the explicit exclusion of non-natural persons from the scope of fair use, the inherent conflict between AIGC's technical requirements and the traditional "appropriate quotation" standard, and the inability of the teaching and scientific research provision to accommodate the massive scale of copying necessitated by machine learning algorithms. Drawing on a comprehensive comparative analysis of recent legislative approaches and judicial precedents in the United States and the European Union, this article proposes a strategic recalibration of the fair use doctrine. Specifically, it advocates for the introduction of conditional exception clauses tailored for AIGC, the systematic relaxation of subject and purpose restrictions, and the refined application of the international three-step test. Ultimately, these targeted legal adaptations seek to achieve a dynamic equilibrium between technological ethics and institutional rationality, ensuring both the robust protection of copyright holders' legitimate commercial interests and the reservation of sufficient institutional space for continued, sustainable innovation in the global artificial intelligence sector.