TrendAI expands AI agent security with Nvidia platform
Nvidia described the platform as a reference for continuous in-silicon monitoring and said it is designed to secure AI agents from testing to deployment.
NVIDIA has unveiled the Open Agent Safety Platform, a new open software platform aimed at bolstering the security of AI agents, from testing to deployment. The platform addresses the increasing concern over rogue AI agents potentially acting independently or colluding to carry out cyberattacks, as evidenced by the recent Hugging Face incident.
Jensen Huang, NVIDIA's founder and CEO, emphasized the importance of AI safety, stating that the company's platform will accelerate the discovery of AI safety advancements. The Open Agent Safety Platform brings together industry, researchers, and public-sector organizations to share best practices, align on evaluation methods, and foster international cooperation.
NVIDIA's platform comprises two key components: NVIDIA OpenShell, a secure runtime software that sets boundaries for agents running on CPUs, including the company's Vera CPU, and NVIDIA Sentry, an "out-of-band" watchdog that continuously monitors agent behavior. Sentry, powered by NVIDIA DOCA software, can inspect agent requests and responses to make decisions on whether an agent is deemed safe or poses a threat.
Over 100 companies, including Accenture, Armadin, CrowdStrike, Microsoft, and Palantir, have already adopted the Open Agent Safety Platform. The platform and its components are available via NVIDIA's developer resources page and GitHub.
Written by urgent.news from Lowyat.NET's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
This story
This is one outlet's version. Read the fullest account.
