Legal Regulation Of Automated Decision-making In The Context Of Algorithmic Governanc

In the era of artificial intelligence,algorithm is no longer defined as a simple mathematical logic,but exists as the underlying technology of artificial intelligence.While the algorithm ADAPTS to our life,some endogenous technological risks continue to emerge,thus generating certain derivative application risks,gradually bringing this emerging technology into the vision of legal regulation.Therefore,based on the theoretical background of algorithm-based governance,this paper will introduce and analyze the automatic decision in detail,and put forward an attempt from the perspective of “selection algorithm” to build a whole-process automation application model and realize the constraints of automatic decision by different automation applications,so as to provide a new idea and path for the legal regulation of embedded automation decision.The first part is the introduction.Firstly,this paper briefly summarizes the extensive and in-depth application of automatic decision making in all aspects of social life and presents the scene of “algorithm selection”,and points out that the selected environment is full of various technical crises.Therefore,the author puts forward the possibility of changing into “selection algorithm”;Then the main line is the operation logic of the algorithm,including the review of the algorithm in advance supervision,the configuration of the algorithm in the process of guarantee,and the interpretation of the algorithm in the post accountability.The current academic literature on the regulation of automatic decision-making under the background of algorithm governance is sorted out and analyzed.The second part is the introduction of algorithm automatic decision.The importance of algorithms as “central nerves” to automated decision making is self-evident.As a series of instructions to solve problems,algorithm contains profound significance beyond technology,which can be understood from two different levels: technical significance and social significance.At the same time,it is necessary to have a clear understanding of other concepts related to automatic decision making algorithm,so as to lay a certain foundation for clarifying the applicable scope of relevant norms.On the basis of the preliminary understanding of a series of related concepts,we have the essential difference between algorithmic automated decision making and several other AI related technical concepts: As an independent and deeply connected concept with other technologies,it has its intrinsic characteristics,such as dynamic,global and uncertainty,and these inevitable attributes will also show a series of derivative characteristics after giving significance to the scene,including the relevance and specificity of the result.In addition,the automatic decision has its unique technical logic,and the analysis of its function allocation and scene judgment will provide the infrastructure for the subsequent technical regulation.The third part combs out the legal risks caused by automatic decision of algorithm.In general,the automatic decision of algorithm will cause both endogenous technology risk and derivative application risk.Among them,the risks of endogenous technology are mainly manifested in that the decision result of the algorithm ignores causality,which will lead to the attribution error of legal liability.As the product of mathematical logic,it also makes the legal discretion absolute;The misunderstanding of algorithm optimization also increases the illegal risk.As for the derivative application risk,it mainly affects the right of equality and freedom in the impact on private rights.In addition,in the alienation of public power,algorithmic decision not only has an impact on the power itself in essence,but also will compress the formal procedure,which will cause certain damage to both substantive justice and procedural justice.The fourth part will analyze the current situation and difficulties of legal regulation of automatic decision-making of algorithm.Firstly,it combs the current relevant laws and regulations directly and indirectly.Secondly,it points out the existing difficulties in regulation,including the uncertainty of regulation scope and higher requirements on regulation ability.The regulation of automatic decision-making not only requires the technical level,but also may need to understand the correlation characteristics with other social factors.In the fifth part,some suggestions are given to improve the regulation of algorithm automatic decision.Through the elaboration of the relevant measures of other countries and international organizations,it is suggested that our country should realize the two-way governance of technology and law,and try to establish the legal regulation system of automation decision-making of embedded algorithms by corresponding to relevant legal norms,so as to achieve the constraint of automatic decision-making of algorithms.External function allocation one-way guidance,internal algorithmic automatic decision-making technology logic as the main line,as far as possible to make law and technology integration,jointly achieve comprehensive rule of law.