This is according to new research by the University of MontanaChatGPT, an artificial intelligence chatbot can match the top 1 percent of human thinkers on a standard test of creativity.
Developed using GPT-4, the AI application excelled in fluency and originality in the Torrance Tests of Creative Thinking, a widely recognized tool for assessing creativity.
For the study, they collected responses from 24 students at the University of Montana. The scores were compared to 2,700 students nationally who took the TTCT. All submissions were reviewed by Scholastic Testing Service, which was unaware of AI being involved. The scores were then compared to 2,700 students nationally who took the TTCT in 2016.
The study said the AI responses were just as creative as the responses of the most creative real people who took the test. It said ChatGPT outperformed a majority of students nationally.
“For ChatGPT and GPT-4, we’ve shown for the first time that it performs in the top 1% for originality,” said Dr. Erik Guzik, an assistant clinical professor in U-M’s College of Business who led the study. “That was new.”
The researchers carefully awaited the results and did not jump to conclusions.
“But we have shared strong evidence that AI appears to be developing a creative capacity similar to or even greater than that of humans,” said Dr. Guzik.
The TTCT contains several assessments: one verbal and one figurative. Both measure divergent thinking, or the thought process used to generate creative ideas, CNBC reported.
In the study, ChatGPT responses were among the highest percentiles for fluency and originality.
“We were all exploring with ChatGPT and we noticed it did some interesting things that we weren’t expecting,” said Dr. Guzik. “Some of the responses were new and surprising. Then we decided to put it to the test to see how creative it really is.”
Featured video of the day
Jamia Millia Islamia will soon start Medical College: Vice Chancellor Najma Akhtar