Legal Problematics of Criminal Liability for Sexual Deepfakes as a Form of Digital Sexual Violence

Abstract

Artificial intelligence (AI) enables the creation of realistic synthetic sexual content using individuals’ faces, bodies, voices, or identities without their consent. Sexual deepfakes raise significant legal concerns because individuals can become victims of digital sexual violence without having created or participated in the depicted content. This study analyzes the legal status of sexual deepfakes as a form of digital sexual violence and examines the criminal liability of their creators and disseminators under Indonesian criminal law. It employs a normative legal research method incorporating statutory, conceptual, case, and comparative approaches. The findings indicate that several Indonesian legal instruments partially address sexual deepfakes, including the Law on Sexual Violence Crimes (UU TPKS), the Electronic Information and Transactions Law (UU ITE), the Pornography Law, the Personal Data Protection Law, and the National Criminal Code (KUHP). However, these instruments do not establish a specific and comprehensive criminal offense encompassing the creation and dissemination of non-consensual synthetic sexual content. The principal legal challenge concerns liability at the creation stage because the content does not constitute an actual recording of the victim. These findings highlight the need for a criminal liability framework that distinguishes among the creation of sexual deepfakes, the use of victims’ identities, and the dissemination of such content. This distinction provides a conceptual basis for addressing gaps in criminal liability while positioning the victim’s lack of consent as the central consideration in legal protection.