default author photo

Tae Joon Jun

University of Ulsan

SCHOLARLY PAPERS

1

DOWNLOADS

94

TOTAL CITATIONS

0

Scholarly Papers (1)

1.

Manipulating Medical Advice Through Stealth Prompt Injection in Large Language Models: An Experimental Study on Vulnerabilities and Patient Safety Risks

Number of pages: 33 Posted: 10 Jun 2025
Inha University Hospital, University of Ulsan, Seoul National University - Seoul National University Hospital, Seoul National University, University of Ulsan - Department of Pulmonary and Critical Care Medicine and Seoul National University
Downloads 94 (720,108)

Abstract:

Loading...

Large Language Models, Prompt Injection, Patient Safety, Cybersecurity, AI Safety