Artificial intelligence (AI) has reached new heights, as per research conducted by the University of Montana and its partners. The study reveals that ChatGPT can now compete with the top 1% of human thinkers in a standard creativity test. Dr. Erik Guzik, an assistant clinical professor at UM’s College of Business, led the study, utilizing the Torrance Tests of Creative Thinking (TTCT), a well-established tool used for decades to assess human creativity.
Also Read: Meta Launches ‘Human-Like’ Designer AI for Images
Testing AI Creativity
In this remarkable experiment, researchers pitted ChatGPT, an application powered by the advanced GPT-4 AI engine, against a control group of 24 University of Montana students enrolled in Dr. Guzik’s entrepreneurship and personal finance classes. These responses were compared with those of 2,700 college students nationwide who took the TTCT in 2016. The Scholastic Testing Service carried out the scoring remaining unaware of the involvement of AI.
Also Read: AI-Generated Content Can Put Developers at Risk
ChatGPT’s Rise to Creative Excellence
The results were awe-inspiring, placing ChatGPT in the ranks of the creative elite. The AI application demonstrated extraordinary fluency, excelling in generating many ideas, and showcased remarkable originality in producing new and fresh ideas. While it slipped slightly to the 97th percentile in flexibility, which measures the ability to generate different types and categories of ideas, ChatGPT’s outstanding performance was undeniable.
Also Read: How to Use Generative AI to Create Beautiful Pictures for Free?
Dr. Guzik’s Insights
Dr. Erik Guzik, the lead researcher, expressed his satisfaction with the study’s outcomes. He noted that ChatGPT’s performance in originality, ranking within the top 1%, was particularly remarkable. While a few of his University of Montana students also achieved this distinction, ChatGPT outshined the majority of college students across the nation.
Conference Presentation and Interpretation
During the spring semester, Dr. Guzik and his team tested the AI and his students. Assisting him in this pioneering work was Christian Gilde of UM Western and Christian Byrge of Vilnius University. In May, the researchers presented their findings at the prestigious Southern Oregon University Creativity Conference. Dr. Guzik emphasized that they refrained from providing extensive interpretations at the conference, instead focusing on presenting the results. However, they did share compelling evidence suggesting that AI’s creative ability appears to be on par with, or potentially surpassing, human capability.
Also Read: Revolutionizing Computer Graphics: NVIDIA to Unveil 20 AI Research Papers at SIGGRAPH 2023
Unveiling the TTCT Test
The TTCT test itself encompasses prompts that mimic real-life creative tasks. For example, participants are challenged to devise new uses for a given product or to improve upon it. Dr. Guzik illustrated this with the example of a basketball, where participants must generate as many alternative uses as possible, such as shooting it in a hoop or incorporating it into a display. This exercise encourages individuals to think outside the box and consider unconventional applications, such as repurposing the basketball as a planter or using brick to build structures or as a paperweight. AI’s ability to generate novel ideas within this framework is nothing short of impressive.
Surpassing Human Imagination
One particularly surprising outcome of the study was AI’s success in generating original ideas, a hallmark of human imagination. The test evaluators were given lists of common responses for each prompt, including those that were expected to be submitted. However, AI emerged in the top percentile for producing fresh & innovative responses, showcasing its remarkable growth in original thinking. Dr. Guzik referenced previous research conducted on GPT-3, the predecessor to GPT-4, which indicated that ChatGPT did not outperform humans in tasks involving original thinking. However, with the more advanced GPT-4, ChatGPT has undeniably secured its place within the top 1% of all human responses.
Also read: Artificial Intelligence vs. Human Intelligence: Top 7 Differences
Our Say
The boundaries of AI continue to expand, and ChatGPT’s exceptional performance in creativity serves as a testament to the incredible progress made in this field. As AI continues to evolve, it is poised to contribute increasingly to innovative problem-solving, sparking curiosity and revolutionizing various industries. The intersection of human intellect & AI capabilities offers a promising glimpse into a future where AI enhances human creativity. We are about to witness AI pushing the boundaries of what we thought was possible.
Sabreena Basheer is an architect-turned-writer who's passioante about documenting anything that interests her. She's currently exploring the world of AI and Data Science as a Content Manager at Analytics Vidhya.
