Comment on “A Comparison of Artificial Intelligence Platforms in the Utility of Answering Frequently Asked Questions About Carpal Tunnel Syndrome”

We would like to comment on "A comparison of artificial intelligence platforms in the utility of answering frequently asked questions about carpal tunnel syndrome: a cross-sectional study." 1 This study compares the accuracy of responses generated by ChatGPT, Google Gemini, and Bing Copilot to questions about carpal tunnel syndrome using the DISCERN tool and the Journal of the American Medical Association score.However, there are numerous constraints to consider.First, only two raters examined responses, which could result in excessive subjectivity.Even with DISCERN, discrepancies between raters can influence the overall score.Furthermore, DISCERN, a broad health data quality assessment technique, may fail to capture in-depth clinical accuracy or the chatbot's capacity to answer complex inquiries.Another disadvantage is that the study only looked at frequently asked questions about carpal tunnel syndrome, not clinical settings or complex patient cases.This hinders the chatbot from performing in real-world clinical circumstances or tailoring responses to specific patients.Furthermore, using the average DISCERN and Journal of the American Medical Association scores as indicators can obscure significant differences in information delivery dimensions such as completeness, reliability, and understandability.For additional discussion, consider whether each chatbot is prone to presenting outdated information, omitting warnings or cautions, and whether responses can be tailored to the general user or a specific patient population.The influence on current users should also be evaluated.Incorrect data collection can lead to misunderstandings and poor health care decisions.Reanalysis techniques for gaining new insights could include increasing the number of raters to increase scoring reliability, assessing replies by inquiry type (eg, symptom, diagnosis, therapy, and preven-