公共安全视角下深度伪造风险的概念模型与生成机理
作者简介:詹国彬,南京审计大学国家安全学院教授 (江苏南京 211815);陈逸凡,南京审计大学国家安全学院硕士研究生 (江苏南京 211815)。
基金项目:
本文为国家社会科学基金重点项目“数字赋能促进公共服务高质量供给及其实现路径研究”(21AGL032)的阶段性成果
摘要: 数字时代背景下,深度伪造技术如同魅影般在网络空间肆意游荡,凭借其以假乱真的特性对公共治理的真实性、透明度以及公信力构成了前所未有的挑战,同时也对公共安全乃至国家安全造成了重大威胁。研究发现,深度伪造风险的生成是一个多机制协同作用的动态过程:技术资本化是起点,驱动深度伪造技术的快速商业化应用;资本权力化助推深度伪造技术异化为社会操控工具,通过制造虚假信息影响公众认知;风险催化机制因技术低门槛和平台算法操控而加剧传播;正反馈机制导致虚假内容在传播中不断自我强化;最终借助风险传导机制形成跨领域、多层次的危害放大效应。为此,应从法律规制、政策监管、制度规范、国际合作、公众参与等维度入手,构建一个复合性的风险治理框架,以规范和推动深度伪造技术在公共治理领域的合法应用与有序发展。
Conceptual Model and Generation Mechanism of Deepfake Risks from the Perspective of Grounded Theory
Abstract: In the digital era, deepfake technology pervades cyberspace, undermining the authenticity, transparency, and credibility of public governance through deception and threatening public and national security. Research shows that deepfake risks emerge from a dynamic, multi-mechanism process:technological capitalization kick-starts rapid commercialization; capital's power transformation weaponizes the technology for social manipulation by spreading misinformation; the risk catalysis mechanism is exacerbated by the low technical threshold and the manipulation of platform algorithms; positive feedback loops reinforce false content during diffusion; and finally, the risk transmission mechanism amplifies cross-domain and multi-level harm. Thus, a holistic governance framework integrating legal regulation, policy oversight, institutional standardization, international cooperation, and public participation is essential to ensure the legal and orderly application of deepfake technology in public governance.