Related Experiment Video
Updated: Jun 27, 2025

Augmenting Large Language Models via Vector Embeddings to Improve Domain-Specific Responsiveness
Published on: December 6, 2024
Language-adaptive artificial intelligence: assessing CHATGPT'S answer to frequently asked questions on total hip
Muhammad Talal Ibrahim1, Sarah Ashraf Khaskheli2, Hania Shahzad3
1Department of surgery, Section of Breast Surgery, Aga Khan University Karachi, Pakistan.
Abstract:
ChatGPT is reported to be an acceptable tool to answer a majority of frequently asked patient questions. ChatGPT also converses in other languages including Urdu, which offers immense potential for the education of Pakistani patients. Therefore, this study evaluated ChatGPT's Urdu answers to the ten most frequently asked questions on Total Hip Arthroplasty, which were then rated by an expert. Out of 10 answers in English, 9 (90%) were satisfactory requiring minimal clarification and 1 (10%) was satisfactory requiring moderate clarification. In both Roman and Nastaliq script Urdu, 1 (10%) answer was satisfactory requiring moderate clarification, while 9 (90%) were unsatisfactory requiring substantial clarification. In conclusion, as opposed to ChatGPT English responses, Urdu responses were much less rigorous, generic, and lacked scientific rigor. We have a long way to go before Pakistani patients with limited English language skills could benefit from AI chatbots like ChatGPT.
More Related Videos
07:45The Use of Mixed Reality in Custom-Made Revision Hip Arthroplasty: A First Case Report
Published on: August 4, 2022
09:51The Transition to an Anterior-Based Muscle Sparing Approach Improves Early Postoperative Function but is Associated with a Learning Curve
Published on: September 7, 2022