Homogenizing Effect of a Large Language Model (LLM) on Creative Diversity: An Empirical Comparison of Human and ChatGPT Writing

Generative AI systems, especially Large Language Models (LLMs) like ChatGPT, have recently emerged as significant contributors to creative processes. While LLMs can produce creative content that might be as good as or even better than human creations, their widespread use risks reducing the diversity of creative outputs across groups of people. In the present research, we aimed to quantify this homogenizing effect of LLMS on collective creativity. Across three preregistered studies, we analyzed 2,200 college admissions essays. Using a novel measure—diversity growth rate—we showed that each additional human-written essay contributed more new ideas than each additional GPT-4 essay. This homogenizing effect persisted even after a range of enhancements to the diversity of the GPT-4 writings, including prompt and parameters modifications. Overall, our findings suggest that, despite improvements in individual creativity, the widespread use of LLMs could diminish the collective diversity of ideas.