Revolutionizing Materials Research with ChatGPT.

Researchers at the University of Wisconsin-Madison have developed a technique using OpenAI's ChatGPT chatbot to extract data from scientific literature, reducing the workload of materials scientists by 99%. The technique involves asking the chatbot to review sentences and decide whether they contain relevant data, presenting the information in a table form for human review. The team has also developed a follow-up question sequence to double-check the accuracy of the data set. The technique could allow researchers to pursue projects they previously didn't have the time, money or people-power to undertake.