fides

使用信息流控制保护 AI 智能体 该仓库包含一个 Jupyter notebook，用于配合学术论文《使用信息流控制保护 AI 智能体》。该 notebook 以教程形式组织，引导读者理解论文中介绍的概念。它展示了在智能体任务中确定性执行安全策略的机制的实际实现，并将概念转化为可执行的代码，供读者实验。 设置说明 该 notebook 已使用 Azure OpenAI Chat Completions API 针对 GPT-4o 和 GPT-4.1 进行了测试，但也可以轻松适配为使用 OpenAI 端点。代码通过 .env 配置文件（包含 AZUREENDPOINT、APIVERSION...

Optimizing Agent Planning for Security and Autonomy

Indirect prompt injection attacks threaten AI agents that execute consequential actions, motivating deterministic system-level defenses. Such defenses can provably block unsafe actions by enforcing confidentiality and integrity policies, but currently appear costly: they reduce task completion...

Securing AI Agents with Information-Flow Control

As AI agents become increasingly autonomous and capable, ensuring their security against vulnerabilities such as prompt injection becomes critical. This paper explores the use of information-flow control IFC to provide security guarantees for AI agents. We present a formal model to reason about t...

AIs as Trusted Third Parties

This is a truly fascinating paper: "Trusted Machine Learning Models Unlock Private Inference for Problems Currently Infeasible with Cryptography." The basic idea is that AIs can act as trusted third parties: Abstract: We often interact with untrusted parties. Prioritization of privacy can limit t...