Evaluation and comparison of large language model responses to patient questions after diagnosis of high-risk human papillomavirus infection: an expert-rated digital patient education study
Zhen Hao, Lin Wang, Yue Wu, Min Zhao, Huijian Li · 2026
Read Evaluation and comparison of large language model responses to patient questions after diagnosis of high-risk human papillomavirus infection: an expert-rated digital patient education study by Zhen Hao, Lin Wang, Yue Wu, Min Zhao, Huijian Li (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/PMC13473150.txt.
Description
Hao and colleagues tested in 2026 how well two large language models answer patient questions after a diagnosis of high-risk HPV infection, a virus that can lead to cervical cancer. They posed 15 typical patient questions to ChatGPT and to DeepSeek, collected three answers per question, and had five blinded gynecology experts rate all ninety answers.
ChatGPT scored higher on accuracy, safety, and completeness, while clarity was similar for both models. About 93 percent of ChatGPT answers had no major error or potential harm, against about 84 percent for DeepSeek, though that gap was not statistically significant. Risky answers clustered on certain topics, including virus subtypes, partner management, and pregnancy. The authors conclude both models answer generally well but still need guideline-based clinical oversight.
Catalog record
| CREATOR | Zhen Hao, Lin Wang, Yue Wu, Min Zhao, Huijian Li |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 45 KB |
| LICENSE | https://creativecommons.org/licenses/by/4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |