Post written by Daniel Yan Zheng Lim, MRCP, MTech, from Duke-NUS Medical School, Department of Gastroenterology and Hepatology, Singapore General Hospital, and Data Science and Artificial Intelligence Laboratory, Singapore General Hospital, Singapore.

This study examines the ability of ChatGPT to answer common questions about endoscopic procedures that patients might pose.

ChatGPT is an advanced artificial intelligence chatbot based on the Generative Pretrained Transformer 3 (GPT3; OpenAI, San Francisco, Calif, USA). It has excellent general language capabilities and is readily accessible to the general public, including patients.

However, it is not specifically trained for medical tasks and can supply incorrect information (ie, hallucinations). It is important for endoscopists to understand the range of reliable information and misinformation that a patient could obtain from consulting ChatGPT.

We found that advice dispensed by ChatGPT was easily comprehended and would be accessible to a layperson. Information pertaining to periprocedural care and EGD, colonoscopy, and EUS (mainly for diagnostic indications) was of fair quality, with minor inconsistencies.

Yet, information on ERCP had significant inaccuracies and errors. Endoscopists must warn patients of the risks of misinformation when consulting ChatGPT. If ChatGPT is used in clinical care as a chatbot, techniques to curb misinformation (eg, fine-tuning, retrieval-augmented generation) will need to be applied.

Visit iGIE’s Facebook, Twitter, and YouTube accounts for more content from the ASGE peer-reviewed journal that launched in December 2022.

Read the full article online.

The information presented in Endoscopedia reflects the opinions of the authors and does not represent the position of the American Society for Gastrointestinal Endoscopy (ASGE). ASGE expressly disclaims any warranties or guarantees, expressed or implied, and is not liable for damages of any kind in connection with the material, information, or procedures set forth.