expressed opinion entrepreneur Contributors are themselves.
You are reading Entrepreneur India, the international franchise of Entrepreneur Media.
The various use cases for OpenAI’s chatbot, ChatGPT, amazed people. One of the areas where chatbots are proving to be a game-changer is education. ChatGPT seems to be able to do a good job in medical education. In a research paper, ChatGPT passed the United States Medical Licensing Examination (USMLE), the exam that medical students take to gain licensure.
Photo by Markus Winkler on Unsplash
In fact, the paper shows that the AI tool passed the exam, including three steps for different levels of medical specialization, and provides insights into the methodology followed to arrive at the answers. The first paper, published in December, showed that ChatGPT was more than 50% accurate across all exams, and is available to read on medRxivops. “ChatGPT exhibits high levels of consistency and insight in its interpretations,” wrote the researchers who wrote the paper. “These results suggest that large language models may aid medical education and possibly clinical decision-making.”
As part of the test, the researchers relied on 376 questions that were made public in the June 2022 sample exam. They also rely on “random spot checks” to ensure that “no answer, explanation, or relevant content is indexed by Google before January 1, 2022.”
Currently, the dataset for the AI tool has been updated to 2021, as pointed out by the company OpenAI. Interestingly, ChatGPT even successfully passed the MBA exam designed by a professor from the Wharton School of the University of Pennsylvania. The chatbot passed the exam with a grade of B and B.
According to the report, the chatbot “does an excellent job of basic operations management and process analysis questions (including case-study-based questions) and explanations.” AI tools are also good at “modifying answers based on human cues.”
