跳到正文
研踪PaperTrail
← 研踪 PaperTrail

arXiv:2606.28666 · v2

Why Trust Your Agent? Empirical Security Gains from TRiSM-Guided Agentic Workflows in Healthcare

首次提交 2026-06-27 09:00
版本更新 2026-11-19 23:07
雷达观测 2026-11-20 14:52
arXiv ↗PDF ↗

摘要

Agent-based AI has enabled the automation of tasks by exposing application tools and resources to large language models (LLMs). However, to improve scope and accuracy, agents are often given access rights that exceed those of ordinary users, introducing significant security risks and potential regulatory breaches. This paper applies the AI Trust, Risk, and Security Management (TRiSM) framework to a medical report-generation application to demonstrate how an insecure baseline workflow can be transformed into a security-conscious agentic workflow. Both designs were evaluated with five LLMs (Claude Haiku 4.5, GPT-4.1-nano, GPT-4.1-mini, GPT-5.4-mini and Gemini 2.5 Flash) across 800 report generations and 500 attack trials covering RAG poisoning, data-field injection and client-side network injection. The redesign reduced attack success from 31% to 10% for RAG poisoning and from 42% to 25% for data-field injection (McNemar p < 0.001 for both), removed network injection by construction, and raised report accuracy from 72.5% to 86.5%. In an exploratory post-hoc grouping by attack goal, success against attacks aimed outside the report task, such as data disclosure and tool abuse, fell from 21.7% to 2.5% and no patient record was disclosed under the redesign. Attacks altering the reported Covid-19 diagnosis remained effective, accounting for 32 of the 35 successful attacks against the redesign. Least-privilege architecture limits what an injection can reach, but not whether a model follows injected content within its task, so clinical outputs need verification independent of the LLM. No model resisted all three attack vectors, so model selection cannot substitute for secure workflow design.

作者备注:19 pages, 6 figures, 8 tables. v2: Substantially revised. Adds threat model and attack-goal analysis; corrects Gemini 2.5 Flash cost figures (output-token price); revises statistical analysis and limitations

AI 论文解读

基于摘要 · 非全文审阅

此版本尚未生成 AI 解读。解读由后台任务生成并按论文版本保存,登录后共享已有结果。

版本差异

本站首次收录为 v2,暂无更早版本可对照。

观测到的版本

相邻记录可能跳过中间版本;首次发现 v2 及以上也会标为版本更新。

v2 · 2026-11-19 23:07 · 最新观测版本

Why Trust Your Agent? Empirical Security Gains from TRiSM-Guided Agentic Workflows in Healthcare

Agent-based AI has enabled the automation of tasks by exposing application tools and resources to large language models (LLMs). However, to improve scope and accuracy, agents are often given access rights that exceed those of ordinary users, introducing significant security risks and potential regulatory breaches. This paper applies the AI Trust, Risk, and Security Management (TRiSM) framework to a medical report-generation application to demonstrate how an insecure baseline workflow can be transformed into a security-conscious agentic workflow. Both designs were evaluated with five LLMs (Claude Haiku 4.5, GPT-4.1-nano, GPT-4.1-mini, GPT-5.4-mini and Gemini 2.5 Flash) across 800 report generations and 500 attack trials covering RAG poisoning, data-field injection and client-side network injection. The redesign reduced attack success from 31% to 10% for RAG poisoning and from 42% to 25% for data-field injection (McNemar p < 0.001 for both), removed network injection by construction, and raised report accuracy from 72.5% to 86.5%. In an exploratory post-hoc grouping by attack goal, success against attacks aimed outside the report task, such as data disclosure and tool abuse, fell from 21.7% to 2.5% and no patient record was disclosed under the redesign. Attacks altering the reported Covid-19 diagnosis remained effective, accounting for 32 of the 35 successful attacks against the redesign. Least-privilege architecture limits what an injection can reach, but not whether a model follows injected content within its task, so clinical outputs need verification independent of the LLM. No model resisted all three attack vectors, so model selection cannot substitute for secure workflow design.

查看此版本解读 →查看此版本原文 ↗