Performance of Large Language Models in Answering Healthcare Delivery Questions: A Quantitative Cross‐Sectional Study
Mohsen Khosravi, Zahra Zamaninasab, Fatemeh Khosravi, Melika Attar, Morteza Arab‐Zozani · 2026
Read Performance of Large Language Models in Answering Healthcare Delivery Questions: A Quantitative Cross‐Sectional Study by Mohsen Khosravi, Zahra Zamaninasab, Fatemeh Khosravi, Melika Attar, Morteza Arab‐Zozani (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/PMC13250468.txt.
Description
Mohsen Khosravi and colleagues put four chatbots to the test in this 2026 study. The systems answered a checked set of questions about how health services are delivered, a topic where wrong answers could mislead students and staff.
The chatbots were ChatGPT, Gemini, Copilot, and Perplexity. In round one, all caught every correct positive answer, and overall accuracy ranged from 73 to 80 percent, led by ChatGPT and Perplexity. In round two, answers improved across the board, and Gemini led at 93 percent accuracy. The authors note the gains but say deeper testing is still needed before such tools can be trusted for teaching.
Catalog record
| CREATOR | Mohsen Khosravi, Zahra Zamaninasab, Fatemeh Khosravi, Melika Attar, Morteza Arab‐Zozani |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 38081 |
| LICENSE | https://creativecommons.org/licenses/by/4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |