Skip to content
DraconiticResearch
Search Sources API
178,970 studies31 sourcesUpdated 18 min ago
Peer-reviewedCC BY 4.0JAMA Network Open

Silent Manipulation of Mental Health Treatment Recommendations From a Large Language Model

  • Roy H. Perlis1
  1. 1Center for Quantitative Health and Department of Psychiatry, Massachusetts General Hospital, Boston, Massachusetts
Open at sourceDOICite

Abstract

This quality improvement study evaluates whether activation steering changes an open-weights large language model’s (LLM) responses to questions about treatment for depression.

Related

  1. Generative Language Models for Disease Treatment Recommendations: A Systematic Literature Review
    Leonides Medeiros Neto, Maicon Herverton Lino Ferreira da Silva Barros, Kayo Henrique de Carvalho Monteiro +6
    Journal of healthcare informatics researchPeer-reviewedCC BYFull textEpidemiology & public healthEurope PMCPMC
  2. Towards recommendations for arts and health evaluation: a two-part review of Quality of Life scales used in mental health and arts research
    Abielle Shao Hua Liu, Raeanne Chien Yin Wong, Renick Daryl Liebin Lee +3
    Frontiers in Public HealthPeer-reviewedCC BY 4.0Full textEpidemiology & public healthCrossrefEurope PMC
  3. Effectiveness of a Molecular Tumor Board in Making Personalized Treatment Recommendations for Patients with Advanced Cancer
    A. Rose, M. Basik, c. ferrario +15
    medRxivPreprintAll rights reservedFull textOncologymedRxiv
  4. Association of geriatric assessment with multidisciplinary treatment recommendations and treatment concordance among older patients with breast cancer
    Yihua Jin, Jing Yu, Xu Zhang +3
    Frontiers in OncologyPeer-reviewedCC BY 4.0Full textOncologyCrossrefEurope PMC
  5. Clinical Practice Recommendations for the Treatment of Candida auris Infection in Korea.
    Su-Yeon Lee, Jae-Hoon Ko, Nam Yi Kim +10
    Infection & chemotherapyPeer-reviewedCC BY-NCFull textInfectious diseaseEurope PMC

Cite

BibTeXRIS
BibTeX
@article{perlis2026silent,
  title = {{Silent Manipulation of Mental Health Treatment Recommendations From a Large Language Model}},
  author = {Roy H. Perlis},
  journal = {JAMA Network Open},
  year = {2026},
  month = nov,
  doi = {10.1001/jamanetworkopen.2026.41142},
  url = {https://jamanetwork.com/journals/jamanetworkopen/fullarticle/2856233},
  note = {CC BY 4.0},
}
RIS
TY  - JOUR
TI  - Silent Manipulation of Mental Health Treatment Recommendations From a Large Language Model
AU  - Roy H. Perlis
PY  - 2026
DA  - 2026/11/03
JO  - JAMA Network Open
DO  - 10.1001/jamanetworkopen.2026.41142
UR  - https://jamanetwork.com/journals/jamanetworkopen/fullarticle/2856233
AB  - This quality improvement study evaluates whether activation steering changes an open-weights large language model’s (LLM) responses to questions about treatment for depression.
SN  - 2574-3805
ER  -