期刊文献+

人工智能“深度伪造”技术风险刑法规制的向度与限度 被引量:13

On the Dimension and Limitation in Criminal Regulation of AI “Deepfake”
原文传递
导出
摘要 "深度伪造"是一种以人工智能深度学习为基础的非真实音视频合成技术。基于"生成对抗网络"之技术内核,"深度伪造"技术所合成的非真实音视频作品具有高逼真度,若被不法利用确实可能引发严重危害后果。但应当看到,"深度伪造"的技术风险在于增加了虚假信息对国家安全、社会秩序或公民权利等传统法益的侵害程度,并未侵犯到新的法益类型。因此,基于法益的立法批判机能,刑法规制"深度伪造"技术的向度应当是在司法层面上降低入罪门槛与提升刑罚量,而并非是在立法层面上盲目地犯罪化。主张以"盗窃个人生物识别信息为坐标将身份盗窃行为入罪化进而规制‘深度伪造’技术"的立法论观点,脱离了立法事实而制造出法益假象,不具有正当性。新兴技术治理过程中,我们要正视刑法的功能限度,刑事高压政策可能会扼杀创新,危及到人工智能技术的发展前景。立足于开放性治理模式并建构以"标识义务"为中心的制度体系,将是"深度伪造"技术法律规制的理性选择。 “Deepfake”is a kind of audio and video processing technology based on the“Generative Adversarial Network”formed by the Deep-learning of big data and artificial intelligence,which can synthesize non-real audio and video works with high fidelity.Of course,“Deepfake”technology is also facing the risk of illegal use.Objectively speaking,the technical risk of“Deepfake”is that it aggravates the infringement of false information on traditional legal interests such as national security,social order or civil rights,but it does not touch the new types of legal interests.Therefore,based on the legislative critical function of legal interest,the direction of criminal law to regulate the technology of“Deepfake”should be to reduce the threshold of conviction and increase the amount of penalty at the judicial level,rather than to adopt the criminal path of blindly adding charges at the legislative level.The legislative viewpoint of adding a new crime with the coordinate of“stealing personal biometric information”so as to regulate the technology of“Deepfake”breaks away from the legislative facts and creates the illusion of legal interest,which is not legitimate.In the process of emerging technology governance,we should face up to the function limit of criminal law.Criminal high pressure policy may stifle innovation and endanger the development prospect of artificial intelligence technology.Based on the open governance mode and constructing the system centered on“identification obligation”,it will be a rational choice for the legal regulation of“Deepfake”technology.
作者 姜瀛 Jiang Ying
出处 《南京社会科学》 CSSCI 北大核心 2021年第9期101-109,共9页 Nanjing Journal of Social Sciences
基金 国家社科基金青年项目“刑法立法模式与修改方式研究”(19CFX038) 中央高校基本科研业务费专项资助项目“网络数据爬取行为刑法规制问题研究”(3132021287)的阶段性成果
关键词 人工智能 深度伪造 犯罪化 司法应对 开放治理 AI deepfake criminalization judicial response open governance
  • 相关文献

参考文献11

二级参考文献77

共引文献253

同被引文献143

引证文献13

二级引证文献18

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部