This thesis delves into the intricate relationship between artificial intelligence (AI), copyright law, and video games, focusing on the recent phenomenon of AI-driven non-playable characters (NPCs). It begins by highlighting the immersive nature of story-driven video games and the pivotal role NPCs play in shaping player experiences. With the advent of AI in gaming, particularly through Large Language Models (LLMs) and Deep Learning Approaches (DLA), NPCs have evolved to autonomously generate diverse and realistic dialogue, thereby enhancing the gaming narrative.
The central inquiry revolves around the extent to which European Union (EU) copyright law protects AI-generated content, using the creation of an app titled “Urban Treasure Hunts” by a Smart NPC in the video game “The Matrix Awakens” as a case study. Through a thorough examination of EU copyright law, including the Berne Convention and CJEU Jurisprudence, this thesis argues that determining the eligibility of AI-generated content for copyright protection hinges on factors such as originality/creativity and expression, with a presumption of human authorship.
Key challenges arise due to the difficulty in discerning AI-generated content from human-authored material and the anthropocentric nature of EU copyright law. The thesis contends that while AI contributes to creative processes, it lacks the same degree of personal touch and intent as human creators. Additionally, the expression prerequisite poses challenges as human authors may struggle to identify AI-generated content accurately. The thesis unfolds its argumentation through a structured analysis, including a review of international and EU copyright law, definitions and concepts of AI, and a detailed examination of “The Matrix Awakens” as a case study. Ultimately, it underscores the need for nuanced consideration of AI’s role in creative endeavors within the framework of copyright law.