A research team led by Professor David Cropley at the University of South Australia has revealed significant limitations in the creative capabilities of generative artificial intelligence models, such as ChatGPT. The study, published in the Journal of Creative Behaviour, asserts that these systems can achieve a maximum creativity score of 0.25 on a normalized scale from 0 to 1, positioning their creative output at the level of an average human. This finding underscores the inability of generative AI to match the originality of highly creative professionals across various fields, including writing, music, art, and scriptwriting.
Professor Cropley explained that while generative AI can adeptly mimic creative processes, it fundamentally lacks the capacity to produce genuinely original and effective works. “While AI can mimic creative behaviour – quite convincingly at times – its actual creative capacity is capped at the level of an average human and can never reach professional or expert standards under current design principles,” he stated. This distinction challenges the prevalent misconception that the ability to generate stories, images, or poems equates to true creativity.
According to Cropley, a significant number of individuals perceive AI systems as creative due to the varying levels of human creativity. He noted that about 60% of people fall below average in terms of creativity, which leads many to mistakenly view models like ChatGPT as creative. “Highly creative people will recognise the weaknesses in the generative AI systems,” he added, highlighting a gap in perception regarding AI’s capabilities.
This research marks the first formal quantitative assessment of AI’s creative capacity, grounded in the mechanics of large language models (LLMs). Cropley emphasized that while generative AI can serve as a valuable tool in creative industries, it cannot replace individuals with exceptional creativity. “A skilled writer, artist, or designer can occasionally produce something truly original and effective. An LLM never will. It will always produce something average, and if industries rely too heavily on it, they will end up with formulaic, repetitive work,” he warned.
Furthermore, Cropley asserted that achieving true expert-level AI creativity would necessitate the development of new computational architectures capable of generating ideas that diverge from established patterns found in existing data. This research provides reassurance to creative professionals worried about the potential impacts of automation on their fields, concluding that “this research shows that the world still needs creative humans – perhaps more than ever.”
The implications of these findings are significant as they challenge optimistic narratives surrounding the potential of generative AI in creative sectors. As organizations increasingly integrate AI technologies into their workflows, understanding the limitations of these tools is crucial for maintaining the integrity and originality of creative outputs.
In a rapidly evolving technological landscape, Cropley’s research serves as a reminder that while generative AI can assist in various tasks, it cannot replicate the unique insights and innovations that human creators bring to their work. As the debate over the role of AI in creative industries continues, this study emphasizes the enduring need for human creativity in a world increasingly influenced by automation.
See also
AI Study Reveals 62% Success in Bypassing Chatbot Safety with Poetry Techniques
Imagiyo AI Image Generator Launches Lifetime Subscription for $34.97, 93% Off Until Dec. 14
AI in Aviation: Key Leaders Stress Need for Verification to Prevent Human Oversights
HSBC Partners with Mistral AI to Enhance Banking Operations with Generative AI Tools
OpenAI Launches GPT-5: Enhanced Multimodal Reasoning and Workflow Management
