Criminal Liability for AI-Generated Content: Emerging Legal Challenges and the Need for a Global Regulatory Framework

Abstract

Generative artificial intelligence has changed how people prepare and share digital content. A user can now create or edit text, images, voices and videos within a very short time. These tools are useful in education, business, entertainment and many other areas. However, they can also be misused. For example, a person may use AI to imitate another person's voice, prepare a fake intimate image, spread a false video or support an online fraud. Such incidents raise an important question: who should be held criminally responsible when AI-generated content is used to harm someone? The answer depends on the facts of each case. The person who gives the prompt, the developer who designs the tool, the organisation that uses it and the platform that shares the content may have different roles. This paper examines whether traditional principles of criminal law, including actus reus, mens rea, causation and individual responsibility, can deal with these situations. It follows a doctrinal and comparative method and considers selected legal developments in the European Union, the United States, the United Kingdom and the Council of Europe. A qualitative comparative matrix is used to examine the legal materials. The study finds that existing criminal laws can cover many forms of deliberate misuse, but practical difficulties remain in identifying offenders, proving intention, collecting digital evidence and investigating cases across borders. The paper argues that a balanced legal response should focus on the conduct and responsibility of each actor. It should protect victims without treating every error made by an AI system as a crime.