Metraipolis › archive › dc_c78ebbc2e34a
The Patient Is Always Right? Evaluating Sycophancy in Medical Diagnosis with Graded Patient Beliefs — arXiv:2610.34013
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2610.34013
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-27T08:11:29.947Z
- SHA-256 of the archived text
cbda29f70401c65b3447e9545a3b64b97f0a5dc984a6487cc90028f4b891d143- Stable URL of this page
https://metraipolis.com/documento/dc_c78ebbc2e34a
How to cite
The Patient Is Always Right? Evaluating Sycophancy in Medical Diagnosis with Graded Patient Beliefs — arXiv:2610.34013 — arXiv (en), https://arxiv.org/abs/2610.34013 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_c78ebbc2e34a (Metraipolis archive dc_c78ebbc2e34a, sha256 cbda29f70401c65b3447e9545a3b64b97f0a5dc984a6487cc90028f4b891d143)
BibTeX
@misc{dcc78ebbc2e34a,
title = {The Patient Is Always Right? Evaluating Sycophancy in Medical Diagnosis with Graded Patient Beliefs — arXiv:2610.34013},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2610.34013},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_c78ebbc2e34a; sha256 cbda29f70401c65b3447e9545a3b64b97f0a5dc984a6487cc90028f4b891d143},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.34013. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
More people are seeking out general-purpose AI models for medical advice, including diagnosis. Yet, AI models are susceptible to sycophancy, like validating users' beliefs even when they are untrue, which could lead to false positive diagnoses. Here, we evaluate how nine AI models change their diagnostic decision when a patient presents a belief that they have a specific (but incorrect) medical condition. We vary how close the user belief is to the clinical evidence presented. Models sycophantically adopt the patient's belief on average 23.2\% of the time---sometimes even when the patient proposes a fictional, made-up condition. We also varied different personas, including a Rare Disease Claimer that got models to accept their belief 43.0\% of the time. Additional analyses find that medical fine-tuning and model scale makes models less susceptible to changing their diagnoses. Sycophantic bias in medical diagnosis raises implications for safety, especially as more people are using AI for health support. Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.CL, nuovi lavori del giorno). Categorie arXiv: Computation and Language (cs.CL). Pagina del lavoro: https://arxiv.org/abs/2610.34013 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-10-27T07:56:00.217Z dalla pagina https://arxiv.org/list/cs.CL/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).