ETHOS: Towards a Modular Ethics Framework for Clinical Multi-Agent Systems

ID: 074c995c
原文链接: https://arxiv.org/abs/2608.15424
PDF: https://arxiv.org/pdf/2608.15424
作者: Rakesh Sharma, Sydney Pugh, Cameron Beeche, Pankhuri Singhal, Rachel Wu, Margaret Eby, Jeffrey Duda, James Gee, Kyra O'Brien, Hersh Sagreiya, Marina Serper, Victoria Gershuni, Angela Bradbury, Anurag Verma, Eric Eaton, Kevin B. Johnson, Walter Witschey
日期: 2026-08-15
更新: 2026-08-15
分类: learning
来源类型: paper
标签: clinical-ai, ethics, governance, multi-agent, arxiv
质量评分: 4/5
抓取时间: 2026-08-19T04:47:39Z

中文导读

基于 LLM 的临床多智能体系统（MAS）可整合多模态患者数据支撑越来越复杂的临床决策，但真实医疗场景部署引发安全公平问责透明与患者信任等伦理关切；WHO美国国家医学院FUTURE-AI 联盟等提出的治理原则大多停留在概念层ETHOS（Ethics and Trust through Hierarchical Oversight System）把模块化伦理框架实现为'治理元 agent'，无需改动架构即可接入任意现有多智能体系统，将组织层面的伦理原则从纸面文档转入运行时强制执行

为什么值得关注

临床多智能体系统的治理元 agent：模块化伦理框架，不改架构接入任意 MAS 并运行时执行

关键信息

论文标题: ETHOS: Towards a Modular Ethics Framework for Clinical Multi-Agent Systems
作者: Rakesh Sharma, Sydney Pugh, Cameron Beeche, Pankhuri Singhal, Rachel Wu, Margaret Eby, Jeffrey Duda, James Gee, Kyra O'Brien, Hersh Sagreiya, Marina Serper, Victoria Gershuni, Angela Bradbury, Anurag Verma, Eric Eaton, Kevin B. Johnson, Walter Witschey
arXiv: https://arxiv.org/abs/2608.15424
发布时间: 2026-08-15
arXiv 分类: cs.MA, cs.AI, cs.LG
关联标签: clinical-ai, ethics, governance, multi-agent, arxiv

English Abstract

The rapid adoption of large language models has enabled the development of clinical multi-agent systems (MAS) capable of integrating multimodal patient data and supporting increasingly complex clinical decision-making. However, the deployment of these systems in real-world healthcare settings raises critical ethical concerns related to safety, fairness, accountability, transparency, and patient trust. While numerous organizations, including the World Health Organization, the National Academy of Medicine, and the FUTURE-AI consortium, have proposed ethical frameworks and governance principles for healthcare AI, these efforts remain largely conceptual. To address this challenge, we present ETHOS (Ethics and Trust through Hierarchical Oversight System), a modular ethics framework designed as a governance meta-agent that can be integrated with any existing multi-agent system without requiring changes to its underlying architecture. ETHOS translates stakeholder-informed ethical requirements into executable runtime oversight through a layered governance approach consisting of deterministic checks, contextual reviews, and a final ethics critic. These components continuously evaluate intermediate reasoning steps and final outputs, enabling the system to identify ethical risks, request revisions, or suppress responses that fail predefined safety and trustworthiness criteria. We demonstrate ETHOS within a hepatology clinical decision-support MAS. Results show that ETHOS improves decision reliability by detecting incomplete, inconsistent, or out-of-scope evidence and appropriately increasing abstention when safe recommendations cannot be supported. By embedding ethical governance directly into system operation, ETHOS provides a practical and auditable mechanism for transforming high-level AI ethics principles into deployable safeguards.

English Summary

Clinical multi-agent systems (MAS) built on LLMs integrate multimodal patient data and support complex clinical decision-making, but deployment raises ethical concerns around safety, fairness, accountability, transparency, and patient trust. Governance proposals from WHO, the National Academy of Medicine, and the FUTURE-AI consortium remain largely conceptual. ETHOS (Ethics and Trust through Hierarchical Oversight System) is a modular ethics framework implemented as a governance meta-agent that can be integrated with any existing multi-agent system without requiring architectural changes, moving organizational ethics principles from documents into runtime enforcement.

Obsidian Notes

内容由 opencli arxiv paper 拉取 arXiv 元数据与摘要生成。
中文导读与价值判断均锚定在条目已有摘要、论文摘要、作者、日期与分类信息上；未补充论文摘要之外的实验细节。
metadata source: opencli