Deepfake Technology and Criminal Liability: Challenges for Contemporary Criminal Law

Abstract

The rapid advancement of deepfake technology, driven by artificial intelligence and deep learning algorithms, has created unprecedented challenges for contemporary criminal law. Deepfakes enable the realistic manipulation of images, audio, and video, making it increasingly difficult to distinguish authentic digital content from fabricated material. While this technology offers legitimate applications in entertainment, education, and digital media, its misuse has facilitated a wide range of criminal activities, including fraud, identity theft, defamation, non-consensual intimate image creation, election manipulation, cyber extortion, and the dissemination of disinformation. These developments expose significant weaknesses in existing criminal law, which was primarily designed to address conventional forms of criminal conduct rather than AI-generated digital deception. This article examines the criminal liability arising from the misuse of deepfake technology through a normative juridical approach supported by comparative legal analysis. It analyzes the adequacy of existing criminal law doctrines concerning actus reus, mens rea, causation, accomplice liability, and corporate responsibility, while evaluating emerging regulatory frameworks in selected jurisdictions. The findings indicate that current legal frameworks remain fragmented and insufficient to address the complex distribution of responsibility among developers, platform providers, content creators, and end users. Accordingly, this study proposes a Risk-Based Deepfake Criminal Liability Framework integrating criminal responsibility, platform accountability, technological safeguards, victim protection, and judicial oversight to strengthen criminal law responses while preserving innovation, freedom of expression, and digital rights in the age of artificial intelligence.