A Random Walk Modeling Framework for Boosting the Creativity of Humans and AI

Abstract

In a rapidly changing and highly competitive environment, constant innovation is necessary to stay ahead of the competition. This makes it critically important to understand the process that leads to innovation and to seek ways to enhance creativity. We develop a model of the creative process that combines concepts from psychology and network science. We use this creative network model to explain how certain interventions to the creative process can boost creativity. The creative network framework can also be used to discover new interventions that are likely to be effective; we propose several novel interventions to demonstrate the concept. We demonstrate the effectiveness of these interventions both analytically and by conducting a series of experiments on generative AI models and human subjects. In a world of human-AI collaboration, it is important to boost the creative potential of both AI and humans.

Keywords: Creativity, ChatGPT, Large Language Models, Creativity of Generative AI, Creativity Models, AI