Research On Data Regulation Of Deep Forgery Technology

After cybernetics in the 1940s and connectionism in the 1980 s,the term Deep fake was coined in 2017 as the nickname of the author who posted fake videos on Reddit in the United States,and has since evolved into an alternative term for the technique.The term is a combination of Deep Learning and Fake.As its application branch came into being,and has profound and extensive influence on politics,economy,society and culture.When it comes to Deep fake,the first thing that comes to people's mind is AI face-changing.Earlier,its applications such as "Fake App","Deep Face Lab" and "Zao" have attracted great social attention and traffic support since their launch.In addition to face changing,Deep fake has many applications,including speech simulation,text generation and micro-expression synthesis.Although deep fake plays a powerful data processing function of its algorithm in the digital era of artificial intelligence,its potential risks also bring huge hidden dangers and threats to citizens' personal information security,social order and national security,as well as great challenges to legal governance and supervision.At present,our country's law has increased the supervision system of artificial intelligence technology,it develops some of laws and regulations,but because of the law contents,abstract is too general and not according to the technical characteristics of Deep fake itself,as well as the data source as a starting point for targeted regulation,the role of law still has limitations.Deep fake is an important achievement of deep learning and big data technology,and data is the source of its technical operation.However,data governance has some obstacles to the current status of Deep fake regulation,mainly involving the technology itself,data rights subject,data application platform,and lack of supervision and other problems.Therefore,for the benign development of Deep fake,we should take data as the tangent point for effective governance.At the same time,the innovation and development of artificial intelligence technology can not be completely suppressed.On the basis of fully considering the regulation of Deep fake data,it should be balanced between technology application innovation and security risk taking.Based on the existing research results,this paper proposes the data regulation framework of Deep fake in China.First is the Deep fake introduce and analyze the basic connotation,and on the basis of the Deep fake characterize the data and the analysis of existing problems,followed by combing the depth of the European Union and the United States forged regulations and practice,to carry on the beneficial reference and the final forging technology for depth data to design a multivariate optimization path of the rule of rules and regulations,Give full play to the effective role of each subject in each link of Deep fake data processing.