Eduardo Bruera, Medical Doctor at MD Anderson Cancer Center, shared on LinkedIn:
“A study comparing ChatGPT, Bard, and Bing on palliative care terms found ChatGPT most accurate (9/10) and comprehensive (8.5/10).
Credibility was low (3/10) across all. Readability was poor (FRE 41.7; grade 14.1). Clinician oversight is needed to avoid misinformation.”
Title: Performance of Three Conversational Artificial Intelligence Agents in Defining End-of-Life Care Terms
Journal: Journal of Palliative Medicine
Authors: Sonal Admane, Min Ji Kim, Akhila Reddy, Michael Tang, Yuchieh Kathryn Chang, Kao-Swi Karina Shih, Maxine De La Cruz, Sammuel Jumary Cepeda, Eduardo Bruera, David Hui
More from Eduardo Bruera on OncoDaily.