e-ISSN 2148-0532 DOI 10.54233 Open Access · CC BY-NC-SA 4.0 Türkçe
Endoüroloji Bülteni — Endourology Bulletin Submit Manuscript
Original Article Open Access

A Comparison Study of AI Response Quality in Kidney Stone-Related Questions: Google vs. ChatGPT

Böbrek Taşıyla İlgili Sorularda Yapay Zekâ (ChatGPT) ve Google Cevap Kalitesinin Karşılaştırılması

Issue
Volume 18 · Issue 1
Published
2026

Abstract

Objective: This study intends to evaluate, in Türkiye during the previous 12 months, the readability, reliability, and general quality of responses given by Google and ChatGPT on often-requested inquiries about kidney stones.

Material And Methods: The 10 most often searched queries and 4 trending topics on kidney stones were found by means of Google Trends data. Fourteen queries in all were asked on Google and ChatGPT. Gunning Fog Index (readability), a modified 5-question DISCERN Score (reliability), the Global Quality Score, and expert urologists’ assessments guided responses’ analysis. Independent t-tests let one evaluate Google’s and ChatGPT’s performance differences.

Results: ChatGPT showed notably better performance than Google in terms of the DISCERN Score (mean: 21.4 vs. 18.0; p = 0.002), Global Quality Score (mean: 4.8 vs. 4.0; p = 0.001), and expert urologists’ ratings (mean: 4.6 vs. 3.8; p= 0.001). Though ChatGPT had a lower Gunning Fog Index (mean: 15.84 vs. 18.23), indicating more legible text, this difference was not statistically significant (p = 0.165).

Conclusion: ChatGPT exceeded Google in offering consistent, high-quality, expert-endorsed knowledge about kidney stones. These results show the possibilities of artificial intelligence language models in access to correct healthcare information and patient education. Additionally, improving transparency of information sources and implementing further verification mechanisms for clinical accuracy are recommended to ensure the reliability of AI-generated medical information.

More From This Issue