Home // AIHealth 2024, The First International Conference on AI-Health // View article

ChatGPT's Accuracy in Answering the National Medical Licensing Examination in Japan

Authors:
Takayuki Nakano

Keywords: generative AI; ChatGPT; Japanese; Medical Licesing Examination.

Abstract:
There is only a few findings about healthcare knowledge belonging generative AI, especially in other than English. We evaluated ChatGPT's accuracy on the Japanese Medical Licensing Examination without any modification of its sentence. ChatGPT (-4) achieved accuracy good enough to pass the exam, as long as questions does not contain images. ChatGPT or other generative AI may have sufficient medical knowledge to apply to healthcare field even in non-English languages.

Pages: 47 to 51

Copyright: Copyright (c) IARIA, 2024

Publication date: March 10, 2024

Published in: conference

ISBN: 978-1-68558-136-7

Location: Athens, Greece

Dates: from March 10, 2024 to March 14, 2024