Legal Regulation Of Algorithmic Discrimination

With the rapid improvement of the level of information technology,human society is gradually entering the era of artificial intelligence.Automated decision-making systems represented by machine learning algorithms have been widely used in various industries.The resulting algorithm discrimination problem has also become the focus of academic circles.Focus.In the era of big data,massive amounts of data provide abundant raw materials for the high-speed operation of enterprises and government agencies.However,the hidden danger that follows is that the discriminatory results generated by the algorithm system may bring about a crisis that is difficult to deal with.Starting from the concept of algorithmic discrimination,this paper analyzes the theoretical basis of its legal regulation and the experience and shortcomings of existing domestic and foreign regulatory methods,and draws an algorithmic governance model that adapts to my country’s national conditions.The full text is divided into four parts.The first part made a systematic legal understanding of algorithm discrimination.Since the academic circles have not uniformly defined the meaning of algorithmic discrimination,this compound term can be disassembled to understand it as a discriminatory situation generated by the algorithmic automated decision-making system,and it can be further based on the existence of discrimination intent and the external manifestation of discrimination These two ways to classify it.The causes of algorithm discrimination can be summarized into three types:one is the limitation of the data set,the other is the effect of human factors,and the third is the product of the algorithm black box.When analyzing and comparing algorithmic discrimination from various sources,it is not difficult to find that they have four common characteristics,namely,high degree of concealment,result magnification,individual difference,and non-traceability.The second part discusses the legal basis of governance algorithm discrimination from the two aspects of necessity and feasibility.With the rapid development of algorithm technology and the increasing popularity of algorithm applications,people’s production and life have been facilitated,and a series of practical problems caused by algorithms have also appeared.Algorithmic discrimination is essentially an extension of existing social discrimination in the computer field.Therefore,the wide application of algorithms will solidify these discrimination concepts in various industries to a certain extent.When the automated decision-making system is introduced into judicial trial work,the resulting discrimination will not only infringe the legal rights of citizens,but also damage the authority of national laws.On the other hand,the iterative development of science and technology law shows that the law has an indispensable regulatory role for the application of technology.For the legal regulation in the algorithm field,not only the theoretical system of computational law can provide a reliable support for the governance of algorithm discrimination,the continuous improvement of personal data rights also increases the possibility of curbing algorithm discrimination from the source.The third part analyzes the status quo of discrimination in governance algorithms at home and abroad.The United States chooses an approach centered on algorithmic responsibility,adopts a multi-field,multi-level decentralized legislative approach,and emphasizes the responsibilities that enterprises should bear in terms of data security.The EU chose a unified legislative model centered on data protection,and the General Data Protection Regulations comprehensively regulate the rights and obligations of enterprises in the control and processing of personal data.Although my country’s legislation on algorithms started late,there are already many laws that provide for data protection.In order to further capture the forefront of high-tech technologies on a global scale,my country has issued a number of strategic documents on the development of artificial intelligence in recent years.However,in the practice of exploring algorithmic regulation,we are still facing many difficulties caused by technical barriers,value conflicts and other reasons.The fourth part puts forward four suggestions on the path of my country’s future governance algorithm discrimination.The first is to establish an advanced algorithm regulation concept:to strengthen technical ethics as a guide,and to restrict algorithm power as the center,to reduce the possibility of algorithm discrimination from the root cause.The second is to clarify the key points of algorithmic legal regulation:perfect the personal data protection system,construct "algorithm interpretation power",and expand the defensive rights of individuals in the face of algorithm infringement.The third is to establish an efficient algorithmic supervision system:set up a special algorithm review agency,establish a government-led supervision model,and strengthen the self-discipline mechanism within the industry.Finally,improve the algorithm accountability and relief system:clarify the legal responsibilities of algorithm developers and users,and establish a class action system in algorithm infringement disputes.