This article has been reviewed according to Science X's editorial process and policies. Editors have highlighted the following attributes while ensuring the content's credibility:


reputable news agency


ChatGPT shows poor performance in answering drug-related questions

ChatGPT shows poor performance in answering drug-related questions

ChatGPT provided no response or incomplete or wrong answers to nearly three-quarters of drug-related questions reviewed by pharmacists, according to a study presented at the American Society of Health-System Pharmacists Midyear Clinical Meeting, held from Dec. 3 to 7 in Anaheim, California.

Sara Grossman, Pharm.D., from Long Island University in New York, and colleagues asked the publicly available, free version of ChatGPT 39 questions that had been posed to a drug information service. Two investigators compared ChatGPT responses to benchmark responses based on a literature search.

Of the 39 questions, the researchers reported that ChatGPT provided responses for 72 percent. Just over one-third of ChatGPT responses (36 percent) were accurate and complete and had no irrelevant information, whereas 64 percent had inaccurate or incomplete responses. One-third of the complex questions and 23 percent of the noncomplex questions were answered in a satisfactory manner.

"Health care professionals and should be cautious about using ChatGPT as an authoritative source for medication-related information," Grossman said in a statement. "Anyone who uses ChatGPT for medication-related information should verify the information using trusted ."

More information: Press Release

More Information

Copyright © 2023 HealthDay. All rights reserved.

Citation: ChatGPT shows poor performance in answering drug-related questions (2023, December 12) retrieved 19 April 2024 from
This document is subject to copyright. Apart from any fair dealing for the purpose of private study or research, no part may be reproduced without the written permission. The content is provided for information purposes only.

Explore further

Battle of the AIs in medical research: ChatGPT vs Elicit


Feedback to editors