一个新的关于深度伪造技术的社会风险评估方法  

A New Social Risk Assessment Method for Deepfake Technology

作　　者：马丽霞 高格格[2] 魏云捷 赫然[2] Ma Lixia;Gao Cege;Wei Yunje;He Ran(School of Economics and Management,University of Chinese Academy of Sciences,Beijing 100190;Center for Research on Intelligent Perception and Computing,Institute of Automation,Chinese Academy of Sciences,Beijing 100190;Academy of Mathematics and Systems Science,Chinese Academy of Sciences,Beijing 100190;Center for Forecasting Science,Chinese Academy of Sciences,Beijing 100190)

机构地区：[1]中国科学院大学经济与管理学院,北京100190 [2]中国科学院自动化研究所智能感知与计算研究中心,北京100190 [3]中国科学院数学与系统科学研究院,北京100190 [4]中国科学院预测科学研究中心,北京100190

出　　处：《管理评论》2024年第8期119-132,共14页Management Review

基　　金：国家自然科学基金面上项目(72101197);国家自然科学基金基础科学中心项目(71988101)。

摘　　要：深度合成(deep synthesis)技术作为人工智能(artificial intelligence)商业化应用的新兴领域,正日益受到社会公众的关注。深度伪造(deepfake)技术是深度合成技术的破坏性应用,对国家安全、社会安全、公民个人安全等构成了较大的威胁,我国及欧美发达国家政府、公司和个人均高度重视其进一步的发展演变。针对现有面向深度伪造技术社会风险定量评估方法缺乏及规制体系不足的问题,本文在综合层次分析法和模糊综合评价等方法的基础上,提出了一个新的用于评价深度伪造技术的社会风险的评估方法。依据评估结果,本文进一步提出了关于我国深度伪造技术目前正处于“中高”社会风险阶段的观点,并进而合理推断出:理论技术不成熟、法律监管不健全、产品服务漏洞多是当前我国深度伪造技术面临的主要问题。本文建立的评估指标体系能够为我国深度伪造技术的应用、社会风险研究的进一步深入提供支持,对规制深度伪造技术社会风险提出的建议能够为有关部门的决策提供参考。As a new field of commercial application of artificial intelligence,deep synthesis is attracting more and more public attention.Deepfake technology is a destructive application of deep synthesis,which poses a great threat to national security,social security,and citizens'personal security.Its further development and evolution have been highly valued by governments,companies,and individuals in China,Europe and the United States.Given the lack of quantitative assessment methods and regulation system of a social risk of deepfake technology,this paper,based on the comprehensive analytic hierarchy process and fuzzy comprehensive evaluation,develops a new social risk assessment method to evaluate the social risk of deepfake technology.According to the evaluation results,this paper further puts forward the view that China's deepfake technology is currently in the stage of"medium and high"social risk and then reasonably infers that the main problem with China's Deepfake technology at present lies in:(1)immature theory and technology,(2)unsound legal supervision and(3)loopholes in products and services.The evaluation index system established in this paper can provide support for the application of deepfake technology and further social risk research in China,and the suggestions on regulating the social risk of deepfake technology can provide reference for the decision-making of relevant departments.

关 键 词：深度伪造技术 新兴技术应用 网络内容安全 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]