版权正当性多次受到重大技术变革的考验，对其反思是一个传统又常新的话题。作为人工智能技术由“量变”到“质变”的标志，以ChatGPT为代表的生成式大模型不仅挑战了现行版权法，更是对版权正当性再次提出了质疑，支撑版权正当性的传统基础理论面临新环境适应性的挑战：劳动财产理论的赋权依据和标准“失灵”，人格理论的先决条件难以为继，激励理论的激励愿景无法达成。ChatGPT语境下单一理论不能完成版权正当性的证成任务，应转而寻找一种多元观念指引下的理论集合：劳动财产理论、分配正义理论和社会规划理论分别在权利范围划定、经济利益分配和社会文化建设层面证成了版权的多元价值，为新技术语境下的版权正当性建构了一个相对连贯、逐级递进的理论基础。

Legitimacy theory has been repeatedly tested by major technological changes. Reflection on the legitimacy of copyright is a traditional and often new topic. As a symbol of the transformation of artificial intelligence technology from quantitative to qualitative, large generative models represented by ChatGPT not only challenge current copyright law, but also question the legitimacy of copyright. The traditional foundational theories that support copyright legitimacy are facing challenges in adapting to new environments. In the context of ChatGPT, the empowerment basis and standard of Labor Theory have failed, the prerequisites of Personality Theory are unsustainable, the incentive vision of Utilitarianism can not be achieved. In the context of ChatGPT, as single theory cannot complete the task of proving the legitimacy of copyright, a collection of theories guided by multiple concepts offers new inspiration to copyright legitimacy. Labor Theory, Justice Theory and Social Planning Theory have respectively proven the diverse values of copyright at the levels of defining of the scope of rights, distributing economic benefits and constructing culture and society. This provides a relatively coherent and progressive theoretical foundation for copyright legitimacy in the context of new technology.