Amazon Bedrock AgentCore 推出时间策略强化 AI 代理安全
原标题：Securing AI agents with temporal policies in Amazon Bedrock AgentCore
AI 摘要
AWS 在 Amazon Bedrock AgentCore 中推出了时间策略（temporal policies），这是一种基于代理轨迹的状态ful授权控制，在网关层评估请求与历史事件，以增强 AI 代理的安全性。该策略可强制工具调用顺序、防止数据伪造、限制累积财务风险，并要求高价值操作需人工审批。由于策略在网关外运行，代理无法绕过，从而解决了传统无状态访问控制无法应对的上下文相关风险。
正文节选
Securing AI agents with temporal policies in Amazon Bedrock AgentCore Before AI agents, it was generally sufficient for access controls to treat each action as an independent event. Applications relied on deterministic business logic to enforce whether actions happened in the right order or whether the data was up-to-date. AI agents behave in fundamentally different ways than traditional applications. They decide at runtime which tools to call, with which arguments, and in what order. That flexi