“元规制”下生成式人工智能治理的反思  

Reflections on the Generative Artificial Intelligence Governance under “Meta-Regulation”

作　　者：张焕培 ZHANG Huan-pei(School of Law and Political Sciences,Guangdong Ocean University,Zhanjiang,Guangdong 524088,China)

机构地区：[1]广东海洋大学法政学院,广东湛江524088

出　　处：《张家口职业技术学院学报》2024年第4期17-21,共5页Journal of Zhangjiakou Vocational and Technical College

基　　金：广东省哲学社会科学规划2021年度粤东西北专项立项研究课题“协同治理视野下南海碳汇交易规则法治化研究”(课题编号:GD21YDXZFX01)的阶段性成果。

摘　　要：生成式人工智能的技术失序对现有的风险治理路径提出了挑战。依托“硬法”的治理路径面临解释论困境,借助“软法”治理框架则难以保证治理效果。而元规制路径所展现出的自我规制与外部规制的双重面向,有利于突破当前的治理困局,且基于合目的性、合规律性与合规范性的考量,这一新型路径本身的合理性得以证成。基于此,在元规制路径下,应坚持基于人本主义的人机共生原则,形塑自我规制履行信息的披露机制,探索人工智能领域的监管沙盒制度,进而在创新激励中统筹兼顾、规范治理,实现生成式人工智能技术发展与安全之间的平衡。The disorder in the technology of generative artificial intelligence poses a challenge to the present risk prevention and control.The governing path relying on“hard law”faces the dilemma of interpretation,while the governing framework relying on“soft law”cannot guarantee governing effectiveness.A dual-aspect of self-regulation and external regulation that the path of meta-regulation demonstrates is conducive to breaking through the current governance deadlock.Based on the considerations of purposiveness,normativity,and legality,the rationality of this new path can be proven.Under the path of meta-regulation,it is necessary to adhere to the principle of human-machine symbiosis based on humanism,shape a self-regulation mechanism for information disclosure,and explore the regulatory sandbox system in the field of artificial intelligence.In this way,while coordinating normative governance through innovation and inspiration,a balance between the development and security of generative artificial intelligence technology can be achieved.

关 键 词：生成式人工智能 法律治理 元规制 

分 类 号：DF369[政治法律—宪法学与行政法学]