美国人工智能适配伦理与管控的政府指引与我国适用之启示  

Ethics and Controt of Artificial Intelligence Adaptation in the United States and Implications for China

作　　者：唐淑臣 许林逸 Tang Shuchen;Xu Linyi(School of Law,Tsinghua University,Beijing 100084;School of Law,Beijing Normal University,Beijing 100875)

机构地区：[1]清华大学法学院,北京100084 [2]北京师范大学法学院,北京100875

出　　处：《情报杂志》2023年第1期52-58,共7页Journal of Intelligence

基　　金：2020年度中国政法大学网络法学研究院支持项目“当前社会治理体系下平台责任研究”。

摘　　要：[研究目的]人工智能在发展迅速的过程中,呈现出违背伦理与法治的趋势,美国政府及时采取措施对人工智能发展进行指引,其中的经验值得总结与借鉴。[研究方法]通过对美国政府指引人工智能的理念和实操进行分析,可以发现美国政府在理念层面要求人工智能的发展必须符合美国价值观和法治理念,在实操层面,则对工程实践、系统性能、人机交互、问责治理中的人工智能存在的问题进行解决。[研究结论]通过吸收美国已有的经验,并对其中的误区进行明辨,我国可以在理念层面明确人工智能的伦理与法律准则,在实操层面强化人工智能的可审计性、可追溯性、可解释性及可依赖性,确保人工智能适配伦理与法治。[Research purpose] In the process of rapid development of artificial intelligence, there is a tendency to violate ethics and the rule of law. The US government has taken timely measures to guide the development of artificial intelligence. The experience is worth summarizing and referring to. [Research method] Through the analyze of the concept of artificial intelligence to the United States government guidance and the application, we found that, in the concept level, the U.S. government require the development of artificial intelligence must conform to the American values and law rules, in application level, require it solve the problems existing in engineering practice, performance of the system, human-computer interaction, accountability management about artificial intelligence. [Research conclusion] By absorbing the existing experience in the United States and making clear the ethics and legal rules of artificial intelligence at the concept level, our country can strengthen the auditability, traceability, interpretability, and dependence of artificial intelligence at the practical level, to ensure the ethics and rule of law in artificial intelligence.

关 键 词：人工智能 伦理与管控 可审计性 可依赖性 可解释性 可追溯性 美国 

分 类 号：F49[经济管理—产业经济] D913[政治法律—民商法学]