Synapse
⌘+K
Synapse
PulseExploreJournal ClubResearchersJournals
Instagram
HomeJournal ClubExplore
June 12, 2026Baylor University Medical Center Proceedings

Evaluating the quality of artificial intelligence responses to psoriasis-related clinical and patient questions: a comparative study of ChatGPT, Gemini, and Microsoft Copilot

View Full Paper
Ask AI
Bookmark
Share

Authors

GDGözde Ulutaş DemirbaşEDEsin DiremsizoğluADAbdullah Demirbaş

Discussion

Loading...

Member takes

Overview

Comparative study evaluates AI chatbot responses for psoriasis, suggesting need for physician oversight.

Key Points

  • This study aims to assess the quality of AI chatbot responses to psoriasis-related questions across different domains.
  • Fifty-four psoriasis-related questions were asked to ChatGPT, Gemini, and Microsoft Copilot.
  • Responses were scored by three board-certified dermatologists based on accuracy, completeness, and safety.
  • Scores for different categories included diagnostic, treatment, and patient questions.
  • Gemini scored the highest in treatment questions (8.00 ± 0.00) and significantly outperformed others in patient questions (P < 0.001).
  • Interrater agreement was substantial, with ChatGPT scoring κ = 0.743, Gemini κ = 0.830, and Copilot κ = 0.844.
  • All models showed high clinical safety, but responses varied in completeness and accuracy.

Cite This Study

Demirbaş et al. (2026) studied this question.

synapsesocial.com/papers/6a2ba4778101cf8926f02ee9https://doi.org/10.1080/08998280.2026.2683945
View Full Paper
Ask AI
Bookmark
Share