The Alienation Of Algorithms And Legal Regulation

While algorithms bring convenience to social life,they also bring hidden worries about rights infringement and power impact.The alienation of algorithms affects social fairness and counterparty rights,bringing risks of individual,social,and public governance and abuse of power.However,algorithmic alienation has difficulties such as technical neutrality defense and insufficient information regulation.This paper takes the model of "who governs,what to govern,and how to govern" as the starting point,and takes the practice of arithmetic laws and regulations and relevant legal norms in recent years as the theoretical basis.In terms of "who governs",based on the state’s basic rights protection obligations and risk prevention needs,the administrative regulation of algorithms is legitimate,and the regulatory path of administrative regulation has more advantages than other regulatory paths.In terms of "what to govern",the institutional content of China’s arithmetic regulation model takes algorithm security as the core goal,trust communication as the communication mechanism,and algorithm responsibility as the regulatory approach,and the improvement of relevant governance content is also based on algorithm security assurance,right protection of trust communication and reasonable limitation of algorithm responsibility.In terms of "how to govern",through the institutional content and evaluation of the extraterritorial calculation legal model,taking comparative law research as the perspective and feasibility as the key content,the institutional effectiveness of different regulatory models is investigated and evaluated,and then absorbed or localized in combination with China’s regulatory practice content.By evaluating the institutional content of the arithmetic regulatory model,the presupposition of the arithmetic regulatory model is the regulatory path: post-event accountability combined with individual empowerment to ensure algorithm security,create a collaborative regulatory method of cooperative supervision and parallel self-discipline supervision,and it is necessary to internalize algorithm ethics into a useful attempt to legalize.Based on the improvement of the system of algorithm security,implement algorithm grading based on impact assessment in the pre-prevention stage,divert and regulate different risks and autonomous algorithms,clarify algorithm audit rules in the post-event review stage,and give full play to the cooperative supervision power of third-party institutions;Based on the right to trust communication,it is necessary to further protect the right to know and apply algorithm interpretation in scenarios.Reasonable limitation of liability based on algorithms requires not only clarifying the scope of application and technical due process in algorithm administration,but also limiting the restrictions on algorithms.