Addressing Deepfakes: Harmful AI-generated Content between Criminal Law and Digital Regulation

This book addresses the legal challenges posed by deepfakes-highly deceptive content created using artificial intelligence-by examining how criminal law and other emerging regulatory measures are being used to address the harms generated by deepfakes.

As AI-generated content becomes increasingly realistic and accessible, the risk of misuse grows more pressing. Hence this book firstly identifies and systematises the varied harms associated with deepfakes-from non-consensual pornography and online harassment to political disinformation and reputational damage.

Secondly, it examines the emergent role of criminal law in responding to these harms, highlighting both recent initiatives worldwide and structural limitations. While several jurisdictions are moving toward criminalising certain harmful uses of deepfakes, this book questions whether-and under what circumstances-criminal law is the most appropriate or effective tool. In particular, it examines two case studies: adult non-consensual intimate material and political disinformation.

Thirdly, it analyses emerging regulatory approaches to deepfakes as both AI-generated outputs and harmful online content. It focuses on transparency obligations and content moderation duties-particularly under the EU's Digital Services Act and AI Act-as alternative or complementary responses to criminal law.

Finally, it investigates the intersection between criminal law and digital regulation, arguing how criminal law remains caught between its symbolic role, often seen as the public's preferred solution, and its more instrumental role in defining illegal content and supporting effective regulation shaped by private platforms and providers.

This is the first book in the field that critically examine the challanges and legal responses to deepfakes across both criminal law and digital regulation, offering a comprehensive analysis of the restrictive legal measures used to address the harms generated by deepfakes, using non-consensual pornography and disinformation online as case studies.