The prosecution of deepfakes used as a means of revenge porn : a comparative study

Abstract

As society evolves and new realities emerge, the realm of sexual offences has also seen the rise of new crimes, driven in part by rapid technological advancements. What was once known as Revenge Porn has now given way to AI-generated Deepfake Revenge Porn, a development that brings new technological capabilities to bear on traditional legal concerns. The concepts of Revenge Porn and AI-generated Deepfake Revenge Porn highlight the inadequacies of existing legal frameworks in addressing the fast-paced changes in modern society. Revenge porn, as traditionally understood, involves the non-consensual sharing of real existing explicit content with the intent to harm, embarrass, or seek revenge on a particular person. Deepfake porn refers to the use of artificial intelligence to create or manipulate digital content, often altering existing images or videos to falsely depict a person in a sexualised manner. While deepfakes do not show real events, they are based on the likeness of real individuals. Their creation oftentimes is done non-consensually, the distribution of which results in synonymous effects as to when real intimate images are disseminated. This swiftly metamorphosing practice undermines established norms and sparks a vital reassessment of established legal structures. Whilst Malta legislates against “traditional” revenge porn via Article 208E of the Maltese Criminal Code titled “Non-consensual taking or disclosure of private sexual photographs and films”, deepfakes produced as a means of revenge porn stand to have no specifically catered legislative protection or ad hoc law upon which it is to be prosecuted. Therefore, the necessity is felt to protect the present generation and the victims of the future, which gives rise to the need for legislation which aligns with present-day realities. Therefore, local existing judicial mechanisms appear insufficient for victims of this crime to obtain reparation, indicating a need for an effective approach to address such abuse and provide for its criminalisation, not only to provide justice to victims but also to deter potential perpetrators. To this effect, this dissertation will look into how AI-generated Deepfake Revenge Porn crimes are legally being approached by the United Kingdom and the United States, given that they are at the forefront of legislating against such acts, with the end goal being of recommending a potential outcome which would be fit for our local legislators in approaching the issue in discussion. The study is tackled specifically from a Criminal Law stance and in relation to adults being exposed to this abuse. The first chapter will contain the Introduction, in which I will provide an overview of the issue raised further to explore its relevance in the context depicted whilst also considering the current situation under Maltese Law. The second chapter will contain the Literature Review, in which I will be delving and analysing relevant literature to gain a deeper understanding of how revenge porn via AI-generated deepfakes is conveyed and its effects. The third chapter will entail the Research Methodology, in which I will be employing a comparative analysis between US and UK legislation, jurisprudence, and ongoing cases in connection to the prosecution of such deepfakes used as a means of revenge porn. In the fourth chapter, I outline the Outcomes of Conducted Research conveyed through the comparative analysis. Finally, the fifth chapter will be composed of the Recommendations and Conclusion, which will focus on the recommendations for remedying the lacuna in our law discussed throughout the study in the event of future revisions.