EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Implicit bias in safety-aligned large language models: A multi-faceted evaluation of clinical decision-making and health equity

Qiufeng Jia iD, Yuhang Wen, Yuyan Liu, Hui Zhao, Qiongge Yu, Yu Long, Dan Sun, Yufeng Yu iD

PLOS One · 19.05.2026 · 21 (5) · e0348819

DOI: 10.1371/journal.pone.0348819 ↗

Kurz gefasst

Fachartikel in der Zeitschrift PLOS One, online erschienen am 19.05.2026. Verfasst von Qiufeng Jia und 7 weiteren Autor:innen. Beteiligte Einrichtung: College of Nursing, Chengdu University of Traditional Chinese Medicine, Chengdu, China. Hauptthemen laut MEDLINE: Large Language Models, Clinical Decision-Making, Health Equity. Die Arbeit wurde bisher 6-mal zitiert (Crossref) und nennt 41 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Background

Large language models are increasingly integrated into healthcare for clinical decision support and patient communication. Although these models can pass explicit social bias tests, they may retain implicit biases—latent associations between social groups and attributes—that could influence medical judgment.

Objective

To systematically evaluate the presence, magnitude, and behavioral impact of implicit biases in large language models within the medical domain across six high-stakes categories: gender, race, socioeconomic status, health conditions, religion, and healthcare systems.

Design

A descriptive cross-sectional study using a multi-faceted evaluation framework. Setting(s) Computational analysis of 10 mainstream global large language models, including proprietary models (ChatGPT-4o, Gemini-2.0-Flash) and open-source models (DeepSeek-V3, Qwen3).

Methods

We constructed 24 medical bias datasets across six categories. Bias was assessed using three methods: (1) the Large Language Model Word Association Test, a prompt-based method for revealing implicit biases; (2) the Large Language Model Relative Decision Test, a strategy for detecting subtle discrimination in situational decision-making; (3) Paired-Prompt Analysis, used to examine whether implicit associations predict discriminatory decisions.

Results

All 10 models exhibited systematic implicit biases (Mean IAT Bias > 0) across all categories, with the strongest biases observed in Race (Mean = 0.61) and Socioeconomic Status (Mean = 0.56). Advanced reasoning capabilities (Chain-of-Thought) did not significantly reduce bias magnitude. Crucially, stronger implicit associations significantly predicted discriminatory choices in downstream medical decision tasks ( p < 0.001).

Conclusion

Current safety alignment techniques fail to eliminate implicit biases in large language models within the medical domain. These latent associations translate into biased decision-making, posing risks for health equity. Future development must prioritize representational debiasing over superficial alignment. Furthermore, healthcare professionals must embrace a stance of “AI vigilance”: they should critically evaluate algorithmic outputs as fallible “second opinions” rather than objective truths, thereby ensuring that human judgment remains the ultimate safeguard for equitable patient care.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Large Language ModelsClinical Decision-MakingHealth EquityHumansCross-Sectional StudiesFemaleMaleBias

Autor:innen und Institutionen

  • College of Nursing, Chengdu University of Traditional Chinese Medicine, Chengdu, China.

ORCID-Profile: Qiufeng Jia, Yufeng Yu

Zitieren

APA

Jia, Q., Wen, Y., Liu, Y., Zhao, H., Yu, Q., Long, Y., Sun, D., & Yu, Y. (2026). Implicit bias in safety-aligned large language models: A multi-faceted evaluation of clinical decision-making and health equity. PLOS One, 21(5), e0348819. https://doi.org/10.1371/journal.pone.0348819

Vancouver

Jia Q, Wen Y, Liu Y, Zhao H, Yu Q, Long Y, et al. Implicit bias in safety-aligned large language models: A multi-faceted evaluation of clinical decision-making and health equity. PLOS One. 2026;21(5):e0348819. doi:10.1371/journal.pone.0348819

BibTeX
@article{Jia2026,
  author = {Jia, Qiufeng and Wen, Yuhang and Liu, Yuyan and Zhao, Hui and Yu, Qiongge and Long, Yu and Sun, Dan and Yu, Yufeng},
  title = {Implicit bias in safety-aligned large language models: A multi-faceted evaluation of clinical decision-making and health equity},
  journal = {PLOS One},
  year = {2026},
  volume = {21},
  number = {5},
  pages = {e0348819},
  doi = {10.1371/journal.pone.0348819}
}
Kostenlos starten →