A New Era of AI Agents Demands a New Security Model
Dell, Monday, September 28th, 2026
Dell outlines how the NVIDIA Open Agent Safety Platform on Dell AI Factory enforces AI agent boundaries at runtime, network and silicon layers.
Dell argues that autonomous AI agents can move laterally, access unauthorized data and trigger privileged workflows, risks that prompting and model-level controls cannot address alone.
It describes the NVIDIA Open Agent Safety Platform, which pairs NVIDIA OpenShell open-source runtime governance with the NVIDIA Sentry reference design running on BlueField-4 DPUs and DOCA for out-of-band monitoring and policy enforcement.
Running on Dell AI Factory infrastructure, the stack gives enterprises a validated path to deploy agentic AI with hardware-rooted security and agent-level runtime governance.