딥페이크 기술을 이용한 명예훼손 범죄에 관한 형사법적 검토

This paper examines the challenges posed to the current criminal law system by deepfake technology, which has emerged as a new and potent tool for defamation, infringing upon individual reputation and social trust. By presenting a “manipulated reality” with audiovisual persuasiveness, deepfakes create a level of social risk fundamentally different from that of text-based disinformation. This study conducts an in-depth analysis of the legal issues that arise when applying existing defamation laws—under the Criminal Act and the Act on Promotion of Information and Communications Network Utilization and Information Protection, etc. (the “Network Act”)—to deepfake content. The analysis reveals significant limitations in the current legal framework. Key challenges include the ambiguity in distinguishing “allegation of false facts” from “expression of opinion” in synthetic media, the difficulty in proving the subjective element of “purpose of slander,” and the inherent conflict with freedom of speech. While specific regulations exist for sexual and election-related deepfakes under the Act on Special Cases Concerning the Punishment, etc. of Sexual Crimes and the Public Official Election Act, they fail to cover the regulatory gap for general defamation. Furthermore, relying solely on judicial interpretation is constrained by the principle of legality (nulla poena sine lege). Therefore, this paper argues for a multi-faceted legislative solution that moves beyond mere interpretation. It proposes the establishment of specific criminal provisions tailored to the unique characteristics of deepfakes, the strengthening of platform operator liability with reference to models like the EU's Digital Services Act (DSA), and the implementation of technical measures, such as mandatory watermarking, to ensure transparency.