Hasan R, Megafu M, Guerrero O, Hunt L
A.T. Still University – School of Osteopathic Medicine in Arizona, Chicago , Illinois , USA

Introduction/Objective
Artificial intelligence and machine learning has ushered in a new age of information and has evolved from previous search engines. Artificial intelligence has spread to different sectors including business finance and now into medicine. Artificial intelligence has increased efficiency by speeding up mundane tasks. For this project, we aim to evaluate the responses of systems such as ChatGPT and Gemini to ensure that they make accurate recommendations that are in accordance with the American Academy of orthopedic surgeons guidelines in recommending different curative measures for carpal tunnel syndrome.

Methods
The 2020 AAOS CPG, a widely recognized and respected source, was the basis for determining recommended and non-recommended treatments in this study. ChatGPT and Gemini were queried on 16 treatments based on these guidelines; eight were recommended treatments for Carpal Tunnel Syndrome, and eight were not recommended treatments. These responses were categorized as “Concordance” or “No Concordance” with the AAOS CPGs.

Results
Among the 16 treatments examined, both ChatGPT and Gemini showed concordance with the AAOS CPG for seven out of the eight recommended treatments. However, concordance significantly varied when responses were compared for the not recommended treatments and the varying strengths of CPGs

Conclusion
This study found that when using artificial intelligence and contrasting to the American Academy of Orthopedic Surgeons recommendation, there were some inaccuracies regarding curative measures for carpal tunnel syndrome. Physicians should advise patients about the possibility of encountering inaccurate responses when consulting systems such as ChatGPT and Gemini. Ultimately, patients should consult a physician for a final recommendation when deciding their overall health.