Research On Algorithmic Impact Assessment Of Automated Administrative Decision-making Risk

Artificial intelligence has brought unprecedented changes to administrative methods,and automated administrative decision-making is quite different from traditional administration in terms of manifestation and operation.With its unique advantages,it is widely used in administrative punishment,administrative licensing,information collection,processing and other fields,which greatly improves administrative efficiency,enhances the scientific nature of decision-making,and improves governments’ service at all levels.However,algorithm-centered automated decision-making has inherent shortcomings such as data flaws and technical black boxes,which pose challenges to the basic principles of administrative openness and due process required by traditional laws,and have caused or are causing a series of public risks.In order to prevent the public risks that may arise from automated administrative decision-making,this paper proposes that a set of assessment framework including early warning,supervision and accountability processes should be constructed from the perspective of algorithmic impact assessment.On the one hand,it aims to better protect the legitimate rights and interests of administrative counterparts under digital administration.On the other hand,research is being conducted to establish a regulatory system that conforms to the current application status of algorithms in China,promoting the improvement of the regulatory system for Internet platforms and the sustainable development of the digital economy.This paper is mainly divided into four parts:The first part analyzes the basic issues of automated administrative decision-making and algorithmic impact assessment,summarizes the connotation and definition of automated decision-making and automated administrative decision-making,and discusses the application scenarios of automated administrative decision-making in practice.At the end of part one,the logical relationship between automated administrative decision-making and algorithmic impact assessment is outlined,and the internal relationship between them is analyzed to provide a preliminary understanding of the algorithmic impact assessment system.The second part analyzes the challenges and public risks of automated administrative decision to administrative law caused by the lack of algorithmic impact assessment system.Automated administrative decision-making does not follow the basic principles of administration,which challenges the basic principles of administrative law.Meanwhile,the public risks existing in current application include the weakening of individual information protection,the change and imbalance of "power-right" pattern,and the challenge to basic administration principles.The third part analyzes the comparative advantages of the algorithmic impact assessment system in the regulation of automated administrative decision-making.By sorting out the limitations of the current regulatory situation of automated administrative decision-making in China in terms of thinking,subjects,tools,and so on,as well as the fact that China still belongs to a regulatory model that attaches importance to the regulation of results while ignoring the process,this paper points out the unique governance advantages of the algorithmic impact assessment system.The fourth part examines the experience of algorithmic impact assessment system in countries and regions outside the region,such as the United States,the European Union,Germany,and Poland,providing references for China’s regulatory path to the application of automated administrative decision-making risks.Based on research of risk analysis,regulation status and overseas experience review,this paper proposes four suggestions to improve the algorithmic impact assessment system for automated administrative decision-making risks in China,mainly including clarifying the application scenarios for automated administrative decision-making,the human resource review process in application,and the legal accountability,rights,and obligations after application,covering the entire process of automated administrative decision-making application.