人工智能如何驱动政府循证决策创新?——一个从知识视角出发的探讨

How Can AI Drive Innovation in Evidence-based Decision-making in Government?——A Discussion from the Perspective of Knowledge

循证决策是政府提高公共决策质量、应对时代变革的重要途径。本文从循证决策的底层逻辑——知识视角出发,梳理并探讨了循证决策由传统上基于经验知识进行决策到人工智能驱动下的决策知识生成特点以及循证决策创新机制。研究发现:循证决策的核心基础是各方行动者所拥有的知识,主要包括基于逻辑分析的“理性知识”和基于经验的“直觉知识”,同时科学研究知识亦作为循证决策的重要参考。人工智能驱动下的循证决策体现出人机不断“迭代”互动的协同决策特征,这一过程的机制创新主要体现在政府决策理念的回应性与前瞻性提升、决策权力的标准化与规范化运行、决策范式的实时性与灵活性改善以及决策过程的民主性与公共价值性彰显。同时,应注重循证决策中人工智能应用存在的诸多制约因素,如技术的成熟度与可靠性、决策者的认知与接受程度、政府组织的制度与政策环境、数据的隐私安全与利用的责任问题等。

Evidence-based decision-making is an important way for the government to improve the quality of public decision-making and cope with the changes of The Times.From the perspective of knowledge,which is the underlying logic of evidence-based decision making,this paper reviews and discusses the characteristics of evidence-based decision making from traditional decision making based on empirical knowledge to decision knowledge generation driven by artificial intelligence and the innovation mechanism of evidence-based decision making.It is found that the core basis of evidence-based decision making is the knowledge possessed by various actors,including rational knowledge based on logical analysis and intuitive knowledge based on experience.Evidence-based decision-making driven by artificial intelligence reflects the collaborative decision-making features of continuous“iterative”humancomputer interaction.The innovation mechanism of this process is mainly reflected in the impact of artificial intelligence on the concept of government decision-making,the standardization and normalization of the operation of public decision-making power,the construction of agile evidence-based decision-making paradigm,and the improvement of public participation and democracy in government decision-making.At the same time,attention should be paid to resolving the many challenges existing in the application of AI in evidence-based decisionmaking,such as the maturity and reliability of the technology,the cognition and acceptance of decision-makers,the system and policy environment of government organizations,and the risks of data privacy security and utilization responsibility.