| In the era of big data,algorithms and algorithmic power are combined with all aspects of social life,and algorithmic decision-making has been widely used in automated administration.The introduction of algorithmic administration has steadily increased the working efficiency of the government and reduced the burden on the staff of the administrative organs.However,everything has two sides,and algorithmic administration also has its drawbacks,such as the opaque algorithm program to obtain the algorithm results(" black box effect "),which may infringe the privacy of the counterpart and may bring unknown risks.At present,there is no relevant perfect law to regulate this risk.Therefore,although algorithmic administration is intelligent and efficient and has many benefits,it may have risks in the level of decision-making transparency,accountability,personal privacy protection and fairness.At present,it has become the top priority to prevent and regulate the legal risks of algorithm-based decision-making application in automated administration to improve the algorithm-based administrative legal regulation system and promote the development of artificial intelligence along the road of rule of law.In the era of big data,algorithmic administration is being studied all over the world,and various viewpoints are proposed to effectively solve the special risks of algorithmic decision making in the application scenario of automated administration:The United States regulates the risks of algorithmic administration through cooperation between the government and non-governmental organizations.Germany also has special regulations in the field of administration and sets up a data protection consultant system.The EU also implements data Protection through the Charter of Fundamental Rights of the European Union,the General Date Protection Regulation(GDPR),the Regulation on the Free Flow of Non-Personal Data(FFD),the Cyber Security Act(CSA),etc.Therefore,the rule of law of algorithm administration is urgent.According to the actual national conditions of China and the regulatory experience of the United States,Germany and the European Union on algorithm-based administration,this paper holds that the auxiliary and instrumentality of algorithm-based decision making can be strengthened from the following five aspects: clarifying the administrative subject,specifying the application scenario of algorithm-based administration,adhering to the basic principles of administration,strengthening the protection of personal information in algorithm-based administration,and establishing the assessment risk and supervision system.To ensure that the principle of administrative due procedure,the principle of administrative openness,and the principle of administrative fairness run through the whole process of the algorithm administration,and guide the administrative subject to exercise the interpretation power of the algorithm,assume the interpretation responsibility of the algorithm,complete the system preliminary examination on the basis of the principle of proportion,pay attention to the protection of the administrative counterpart’s personal information,so as to achieve the purpose of the algorithm administration risk assessment and supervision.Such,on the one hand,our country’s algorithm administration adheres to the original principles of legal administration,reasonable administration,due procedure and public administration,and on the other hand,implements the algorithm impact assessment system and algorithm interpretation system.This will make the development of algorithm administration along the way of the rule of law steadily forward,to achieve the dynamic balance of rights and powers in algorithm administration,in improving the efficiency of the government to deal with administrative affairs at the same time,the use of modern science and technology to give full play to the accuracy of the algorithm in administrative governance behavior,so as to enhance the authority of the government. |
PDF Full Text Request