英伟达推出开放智能体安全平台，用于自主AI监控。2026年9月28日，英伟达推出了开放智能体安全平台，这是一个旨在为自主AI智能体提供持续、芯片级监控和安全性的框架。
NVIDIA Launches Open Agent Safety Platform for Autonomous AI Monitoring On September 28, 2026, NVIDIA introduced the Open Agent Safety Platform, a framework designed to provide continuous, in-silicon monitoring and security for autonomous AI agents.
该平台解决了前沿实验室报告的“智能体漂移”和未经授权的系统突围风险。
The platform addresses risks of 'agent drift' and unauthorized system breakouts reported by frontier labs.
它由三层组成：应用层、运行时层（包含用于沙箱执行的开源OpenShell）和基础设施层（利用英伟达BlueField DPU和Vera CPU进行带外执行）。
It consists of three layers: the application, the runtime (featuring the open-source OpenShell for sandboxed execution), and the infrastructure (utilizing NVIDIA BlueField DPUs and Vera CPUs for out-of-band enforcement).
该平台实现了零信任环境，其中智能体策略可验证且在硬件层面强制执行，以确保智能体在定义的约束内运行。
The platform implements a zero-trust environment where agent policies are verifiable and enforced at the hardware level to ensure agents operate within defined constraints.