GPT-4 erreicht in einem weit verbreiteten Kreativitätstest (TTCT) Resultate, die dem obersten Prozent von menschlichen Proband:innen entspricht.

This research seems to be the first to show that AI matches or exceeds human abilities for original thinking. Based on the research, not only are the latest forms of AI generating large numbers of ideas (fluency) and different types, variations, and categories of ideas (flexibility), they are, for the first time, generating new, unique, and unexpected ideas (originality), performing in the top percentile for original thinking. In short, AI models like GPT-4 are becoming capable of producing ideas that humans consider to be original, novel, and unique.

This exploratory research investigated the creative abilities of OpenAI’s large language model, ChatGPT, based on the GPT-4 architecture, as assessed by the Torrance Tests of Creative Thinking. In comparison to human samples and a national percentile from Scholastic Testing Services, ChatGPT’s performance was analyzed for fluency, flexibility, and originality. Results indicated that ChatGPT scored within the top 1% for originality and fluency, and showed high scores for flexibility, thus highlighting the current creative abilities of AI and the potential of AI systems to support and augment human creativity in new and meaningful ways. The study encourages additional research to further define, measure, and develop creativity in the era of advanced AI.