A study on improving in-game user-NPC interaction through prompt engineering using interpersonal relationship type

The current metaverse industry is experiencing a rapid increase in demand due to the development of artificial intelligence(AI) technology and the spread of non-face-to-face lifestyles. However, users do not see much difference between the interaction between non-player characters(NPCs) and users in metaverse games and existing general games. In addition, it is difficult for users to immerse themselves. We proposes a method to improve the interaction between users and NPCs to enhance immersion in metaverse games. This method analyzes real-world human interpersonal patterns and applies prompt engineering techniques to train generative AI. Users can naturally converse with NPCs based on the trained generative AI, providing an experience similar to that of a real person. This method clearly defines the NPC concept for generative AI and minimizes illusions, creating an environment where NPCs can naturally engage users in conversation.