AI-generated clinical summaries require more than accuracy
JAMA Network; by Katherine E. Goodman, JD, PhD; Paul H. Yi, MD; and Daniel J. Morgan, MD, MS; Originally published 1/29/24, redistributed 2/20/24
... Currently, there are no comprehensive standards for LLM-generated [Large Language Model] clinical summaries beyond the general recognition that summaries should be consistently accurate and concise. Yet there are many ways to accurately summarize clinical information. Variations in summary length, organization, and tone could all nudge clinician interpretations and subsequent decisions either intentionally or unintentionally. To illustrate these challenges concretely, we prompted ChatGPT-4 to summarize a small sample of deidentified clinical documents. [Click on the title's link to view the example.]
