NVIDIA Debuts Platform to Secure AI Agents From Testing to Deployment


NVIDIA Debuts Platform to Secure AI Agents From Testing to Deployment

📅 - NVIDIA has introduced an agent security platform that lets operators place enforceable controls around software, including a watchdog designed to stop agents crossing boundaries. Its OpenShell runtime is available; the Sentry watchdog is a reference design tied to BlueField-4 processors. For cloud providers, the launch raises a question: who sets those boundaries and answers when they fail?

That question becomes harder when an agent can use tools, retrieve data and keep working after a user has stepped away. A model may have instructions about what it should do, but a hosting platform needs a way to restrict what the running process can actually reach. NVIDIA's Open Agent Safety Platform [...][... Check source for end of article ...]
Tags: AI

Reads: 3 | Category: General | Source: Hosting Jurnalist : Hosting Jurnalist | Author:
URL source: https://hostingjournalist.com/news/nvidia-debuts-platform-to-secure-ai-agents-from-testing-to-deployment
Want to add a website news or press release ? Just do it, it's free! Use add web hosting news!