On the Tort Liability Determination of Generative Artificial Intelligence—An Empirical Study Based on the “First AI Cases” Nationwide

随着生成式人工智能技术的广泛应用，其在著作权、人格权及不正当竞争等领域引发的侵权争议日益增多，传统侵权责任体系面临挑战。本文以十余起“首例”生成式人工智能侵权司法判例为基础，系统梳理生成式人工智能侵权责任认定的关键标准，包括行为构成、主观过错、因果关系与免责事由，并归纳分析责任主体结构及裁判中适用的法律规范。在此基础上，提出应围绕责任归属、划分与豁免机制，构建以技术介入程度与控制能力为核心的责任认定体系，以实现权利保护与产业发展的有机平衡。With the widespread application of generative artificial intelligence (GAI) technologies, infringement disputes have increasingly emerged in areas such as copyright, personality rights, and unfair competition, posing challenges to the traditional tort liability framework. Based on over a dozen pioneering judicial cases involving GAI-related infringement, this paper systematically examines the key standards for tort liability determination, including conduct identification, subjective fault, causation, and grounds for exemption. It further analyzes the structure of liable parties and the legal provisions applied in relevant rulings. On this basis, the paper proposes constructing a new liability determination system centered on the degree of technological intervention and control capability, aiming to achieve a balance between rights protection and industrial development.