摘要
算法行政作为数字社会的产物,正深刻重塑传统行政的理念、架构与流程。面对算法驱动下行政空间“去疆域化”、行政时序“去周期化”、行政角色“去威权化”与行政法权“去模式化”的颠覆性变革和涌现性行政风险,传统行政风险规制模式预设环境的“稳定性”、奉行“命令—控制”的科层分治体系、秉持“目标考核与事后问责”的线性反馈机制难以为继。在风险评估环节,静态评估范式难以驾驭算法行政的复杂机理;在风险决策环节,线性决策范式难以匹配算法行政的涌现效应;在风险问责环节,刚性问责机制无法契合算法治理的韧性需求;在风险沟通环节,单向传递模式呈现对社会责任共担的结构性排斥。未来的算法行政风险规制应立足反思性评估、包容性决策、语境化问责与建构性沟通,从“稳定性”走向“适应性”。在风险评估领域,应从静态框定走向动态适应,构建复杂性思维范式;在风险决策领域,应从层级管控走向协同治理,实现多元主体的优化博弈;在风险问责领域,应从归责惩戒走向问题解决,在规范性与包容性间实现平衡;在风险沟通领域,应从威权说教走向协商对话,在互信构建中提升规制合法性。
As a product of the digital society,algorithmic administration is profoundly reshaping the concepts,structures,and processes of traditional administrative systems.Confronted with algorithm-driven disruptive transformations,such as the “de-territorialization” of administrative space,the “de-periodization” of administrative time,the “de-authoritarianization” of administrative roles,and the “de-institutionalization” of administrative legal rights,as well as emergent administrative risks,the traditional regulatory model for administrative risks,premised on assumptions of environmental “stability”,hierarchical “command-and-control” systems,and linear feedback mechanisms centered on “performance evaluation and ex post accountability”,has become unsustainable.In the risk assessment stage,the static assessment paradigm fail to address the complex mechanisms of algorithmic administration,In the risk decision-making stage,the linear decision-making framework cannot match the emergent effects of algorithmic governance.In the risk accountability stage,the rigid accountability mechanism clashes with the resilience requirements of algorithmic governance.In the risk communication stage,the one-way dissemination model structurally excludes shared social responsibility.The future regulatory model for algorithmic administrative risks should be based on reflective assessment,inclusive decision-making,contextualized accountability,and constructive communication,transitioning from “stability” to “adaptability”.In the realm of risk assessment,it should shift from static framing to dynamic adaptation,constructing a complexity-oriented thinking paradigm;in risk decision-making,it should evolve from hierarchical control to collaborative governance,achieving optimized interactions among multiple stakeholders;in risk accountability,it should move from blame attribution to problem-solving,balancing normativity with inclusiveness;and in risk communication,it should shift from authoritarian monologues to deliberative dialogues,enhancing regulatory legitimacy through mutual trust building.
作者
张小康
李冰强
Zhang Xiaokang;Li Bingqiang(School of Law,Shanxi University,Taiyuan,Shanxi 030006)
出处
《河南财经政法大学学报》
2025年第3期12-24,共13页
Journal of Henan University of Economics and Law
基金
2023年中国法学会自选课题“人工智能的公法规制研究”[项目编号:CLS(2023)D70]的阶段性成果。
关键词
算法行政
数字政府
风险规制
风险评估
风险沟通
风险问责
algorithmic administration
digital government
risk regulation
risk assessment
risk communication
risk accountability
