智能体安全性有多脆弱？重思意图隐藏与任务复杂度下的智能体风险
How Brittle is Agent Safety? Rethinking Agent Risk under Intent Concealment and Task Complexity
AI总结：
针对大语言模型驱动智能体安全评估忽视复杂任务中隐藏恶意意图的问题，提出二维分析框架并构建OASIS基准，揭示意图隐藏导致安全对齐急剧下降及“复杂度悖论”现象，为强化智能体安全提供基础。
摘要：
当前针对大语言模型驱动智能体的安全性评估主要聚焦于原子性危害，未能应对恶意意图隐藏或稀释在复杂任务中的复杂威胁。我们通过对意图隐藏与任务复杂度这两个正交压力下的智能体安全脆弱性进行二维分析来填补这一空白。为此，我们引入OASIS（正交智能体安全探究套件），这是一个具有细粒度标注和高保真模拟沙箱的分层基准。研究发现两个关键现象：随着意图变得模糊，安全对齐会急剧且可预测地下降；出现“复杂度悖论”，即智能体在更难任务上看似更安全，实则仅因能力限制。通过发布OASIS及其模拟环境，我们为在这些被忽视的维度上探索和强化智能体安全提供了原则性基础。
英文摘要：
Current safety evaluations for LLM-driven agents primarily focus on atomic harms, failing to address sophisticated threats where malicious intent is concealed or diluted within complex tasks. We address this gap with a two-dimensional analysis of agent safety brittleness under the orthogonal pressures of intent concealment and task complexity. To enable this, we introduce OASIS (Orthogonal Agent Safety Inquiry Suite), a hierarchical benchmark with fine-grained annotations and a high-fidelity simulation sandbox. Our findings reveal two critical phenomena: safety alignment degrades sharply and predictably as intent becomes obscured, and a "Complexity Paradox" emerges, where agents seem safer on harder tasks only due to capability limitations. By releasing OASIS and its simulation environment, we provide a principled foundation for probing and strengthening agent safety in these overlooked dimensions.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
