Related Experiment Video
Updated: Jun 19, 2025

Augmenting Large Language Models via Vector Embeddings to Improve Domain-Specific Responsiveness
Published on: December 6, 2024
Disparities in medical recommendations from AI-based chatbots across different countries/regions.
Khanisyah E Gumilar1,2, Birama R Indraprasta3, Yu-Cheng Hsu4,5
1Graduate Institute of Biomedical Science, China Medical University, Taichung, Taiwan. khanisyah@fk.unair.ac.id.
AI chatbot healthcare information varies significantly by region and platform. Bard generally provided superior responses, highlighting the need for equitable access to reliable AI medical advice.
Area of Science:
- Artificial Intelligence in Medicine
- Oncology Decision Support
- Global Health Informatics
Background:
- Healthcare information access is crucial for patient outcomes.
- AI chatbots offer potential for disseminating medical knowledge.
- Geographic and platform-based disparities in AI-generated health information are not well understood.
Purpose of the Study:
- To evaluate AI chatbot performance in providing adjuvant therapy recommendations for endometrial cancer.
- To identify disparities in AI healthcare information quality across different regions and platforms.
- To assess the potential of AI chatbots in delivering equitable medical guidance.
Main Methods:
- Utilized identical case-based questions for AI chatbots (Bard, Bing, ChatGPT-3.5) across four regions (Indonesia, Nigeria, Taiwan, USA).
- Employed a double-blinded evaluation by ten endometrial cancer experts assessing relevance, clarity, depth, focus, and coherence.
- Analyzed response variations based on geographic location and AI platform using statistical methods.
Main Results:
- Significant variations in AI chatbot responses were observed across different countries/regions (p < 0.001).
- Bing chatbots in Nigeria demonstrated superior performance (p < 0.05) across all criteria (p < 0.001).
- Bard chatbots consistently outperformed ChatGPT-3.5 and Bing globally (p < 0.001), with notable strengths in Nigeria (p < 0.05).
Conclusions:
- AI-generated healthcare information quality is influenced by user location and the AI platform used.
- Significant disparities exist, necessitating further research and development for equitable access to trustworthy AI medical information.
- Optimizing AI chatbots for consistent, high-quality medical advice globally is essential.
Related Concept Videos
Issues And Trends In Healthcare Delivery System
Cost Containment
Payment for healthcare services has historically promoted adoption of costly and often unnecessary or inefficient...
Barriers to Effective Communication II
Cultural barriers:
Differences in values, beliefs, religion, knowledge, and tradition can significantly impact communication. Awareness of nonverbal cues is critical, especially when conversing with a patient from a different culture. What appears appropriate in one culture may be inappropriate in another.
Semantic barriers:
As a result of their tendency to use...
The Availability Heuristic
Ethical Issues
Ethical Concerns in Healthcare:
Current Trends in Nursing II
Bioequivalence: Overview

