人工智能在国家治理中的应用及挑战  

Applications and Challenges of Artificial Intelligence in National Governance

作　　者：郭凤林 Guo Fenglin

机构地区：[1]北京外国语大学国际关系学院,100089

出　　处：《世界社会科学》2025年第2期98-115,245,共19页SOCIAL SCIENCES INTERNATIONAL

基　　金：2021年度国家社会科学基金青年项目“新冠肺炎疫情防控中科学-政府-大众关系的社会治理研究”(21CZZ014)的阶段性成果。

摘　　要：当前,人工智能正在突破产业领域,与国家治理相融合,成为国家治理的重要工具,深刻影响着国家治理形态。人工智能是会改善还是会恶化现有的国家治理局面?对全球范围内人工智能在国家治理层面的应用梳理发现,计算智能、感知智能、认知智能三种人工智能形态已在国家治理中得到了初步应用,显著简化了公共行政程序,优化了公共决策效率,拓展了政府回应能力。但在应用过程中,人工智能暴露出数据偏差、算法偏差、国家能力泛化等问题,这对人工智能治理的合法性与安全性提出了挑战。Artificial intelligence(AI)is making breakthroughs in various industrial sectors and increasingly integrating with national governance,profoundly innovating the tools of public administration and the forms of governance.A key question arises:will AI improve or worsen the current state of governance?Based on a global review of AI application in national governance,this paper finds that three forms of AIcomputational intelligence,perceptual intelligence,and cognitive intelligence—have been applied into governance.These technologies have significantly streamlined public administration processes,enhanced public decision-making efficiency,and expanded government responsiveness.However,AI has also revealed several challenges,such as data bias,algorithmic bias,and over-generalization of state capacity.These issues pose new challenges to the legitimacy and security of AI-driven national governance.

关 键 词：人工智能 国家能力 国家治理 科技与政治 

分 类 号：D63[政治法律—政治学]