Author

Ding Xiaodong, Deputy Dean and Professor, Renmin University of China law school.

Abstract

The impact assessment regime is widely used in fields such as environmental protection, privacy and personal information protection and algorithms and artificial intelligence regulation. This system attempts to provide penetrating supervision on professional technical issues or information black boxes, while providing information for supervision and preventing related risks. However, in practice this system also faces issues such as uncertainty in evaluation, formalism in evaluation, and the possibility of hindering development. There are also controversies over specific institutional designs such as whether to assess or not, who to evaluate, when to evaluate, and how to evaluate. The root cause of the problems in the impact assessment regime is that impact assessment is not a purely professional issue in natural sciences or related fields, but a process that involves value judgment and problem solving. Impact assessment should move from domain—specific professional assessment to social legitimacy assessment, from command—and—control assessment to learning governance assessment, and from pure preventive assessment to assessment that considers both safety and development. The necessity of impact assessment, participating subjects, time nodes, questioning and answering modes, and other specific systems should also be optimized.

Keywords:impact assessment regime;risk prevention;environmental protection;personal information protection;artificial intelligence