生成式人工智能失范性风险的法律因应——以凯尔森规范性理论为视角  

Legal Responses to the Anomie Risks of Generative Artificial Intelligence:from the Perspective of Kelsen’s Normative Theory

作　　者：罗艺 陈佳星 LUO Yi;CHEN Jiaxing(Environmental Law College Gansu University of Political Science and Law,Lanzhou,Gansu,China,730030)

机构地区：[1]甘肃政法大学环境法学院,甘肃兰州730030

出　　处：《贵州大学学报（社会科学版）》2023年第5期98-108,共11页Journal of Guizhou University(Social Sciences)

基　　金：2021年甘肃省青年博士基金项目“‘碳中和’目标下智慧生态城乡建设协同发展的法律促进”(2021QB-097);2021年甘肃省高等学校创新基金项目“黄河流域高质量发展的绿色城镇化法律促进研究”(2021B-178)。

摘　　要：以ChatGPT4为代表的生成式人工智能具有高度类人性和广泛适用性,有利于生产方式和众多领域的转型升级。然而,这种所谓强人工智能时代的标志性产物,伴随着技术固有的风险性问题,引发伦理以及法律层面的失范性问题。生成式人工智能的失范性主要表现为:一是针对生成式人工智能引发的风险问题,伦理与法律尚无回应;二是现有的伦理体系和法律制度的“规范性”不足。为此,文章借鉴凯尔森的规范性理论,采用“旧论新用”的方法:一是利用伦理规范化,增强其约束性、可执行力,整理其内部层级关系,形成生成式人工智能的伦理治理体系;二是运用法律规范化,强化其对生成式人工智能数理化逻辑运行的模型规范。最后,基于伦理与法律协同构建生成式人工智能的规范性体系。The artificial intelligence represented by ChatGPT4,which has highly human-like and widely applicable,and is conductive to transformation and upgrading production methods and many fields.However,this iconic product of the so-call strong artificial intelligence era is accompanied by the inherent risks of technology,which leads to ethical and legal issues of anomie.The anomie of generative artificial intelligence is mainly manifested in:first,ethics and law have not yet responded to the risk caused by generative artificial intelligence;second,the existing ethic system and legal system and legal system are not“normative”enough.Therefore,drawing on the experience of Kelsen’s normative theory and adopts a new approach from the old theory,the one is to utilize ethical standardization to enhance its binding character and executable force,arranging its internal hierarchical relationships,forming an ethical governance system for generative artificial intelligence;the second is to apply legal standardization,Strengthen its model specification for the operation of generative artificial intelligence mathematical logic.Finally,building a normative system for generative artificial intelligence based on ethics and law.

关 键 词：生成式人工智能 ChatGPT 失范性 规范性理论 科学化 

分 类 号：DF639[政治法律—刑法学]