生成式人工智能的侵权责任探究

Research on Tort Liability of Generative Artificial Intelligence

摘要: 生成式人工智能的应用带来一系列的侵权风险及侵权责任承担困境，由于生成式人工智能不具备形式逻辑和人类理性，因此生成式人工智能并不具备责任能力。为明确生成式人工智能的侵权责任承担机制，需要划清与生成式人工智能侵权相关主体的责任，现行的网络侵权责任和产品责任的规定并不能完全适用于生成式人工智能侵权，应当调整相关的法律，并针对生成式人工智能侵权建立特殊的责任承担方式，以缓解科技对法律的冲击。

Abstract: The application of generative artificial intelligence brings a series of infringement risks, but due to the problems of multiple infringement subjects, unpredictable infringement behavior, non-specific damage results, complex causal relationships, and difficulty in fault identification and proof, it’s difficult to determine the infringement liability of generative artificial intelligence. Because of the lack of formal logic and human rationality in generative artificial intelligence, it’s not beneficial to advocate for empowering it in order to solve the problem of responsibility taking in generative artificial intelligence. To clarify the tort liability mechanism of generative artificial intelligence, it is necessary to clarify the responsibilities of the parties related to generative artificial intelligence infringement. The current provisions on network tort liability and product liability cannot fully apply to generative artificial intelligence infringement. Relevant laws should be adjusted and special liability bearing methods should be established for generative artificial intelligence infringement to alleviate the impact of technology on the law.