Re: Chen JL et al: Assessment of Correctness, Content Omission, and Risk of Harm in Large Language Model Responses to Ophthalmology Continuing Medical Education Questions
James Vassallo · 2026
Read Re: Chen JL et al: Assessment of Correctness, Content Omission, and Risk of Harm in Large Language Model Responses to Ophthalmology Continuing Medical Education Questions by James Vassallo (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/PMC13396827.txt.
Description
James Vassallo writes a 2026 letter to the editor about a study by Chen and colleagues that graded large language models on eye medicine exam questions for doctors. He praises the study's design, which went past right answers to score omissions and the risk of harm in model replies.
His complaint is timing. The models tested were already old when the paper appeared, and print publishing moves too slowly to keep pace with tools that update every few months. He cites figures from the study itself, including one model at 49.2 percent accuracy with a worrying 44 percent harm risk, and warns readers not to carry verdicts on retired models into judgments about current ones.
Catalog record
| CREATOR | James Vassallo |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 3050 |
| LICENSE | https://creativecommons.org/licenses/by/4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |