- Nvidia’s DOCA Argus promises real-time protection for AI compute environments
- The system runs off-host to detect and respond without performance impact
- Cisco and Nvidia are working to secure scalable AI infrastructure deployments
Nvidia is extending its AI ambitions into cybersecurity with the launch of DOCA Argus, a new software framework aimed at protecting AI infrastructure in real time.
Part of Nvidia’s DOCA software platform, Argus runs on its BlueField networking hardware to detect and respond to threats as they happen, without relying on traditional host-based security tools.
“Cyber defenders need robust tools to effectively protect AI factories, which serve as the foundation for agentic reasoning,” said David Reber, chief security officer at Nvidia. “The DOCA Argus framework delivers real-time security insights to enable autonomous detection and response – equipping defenders with a data advantage through actionable intelligence.”
Built to work with existing enterprise security setups
Argus is designed to operate independently of the system it monitors, avoiding integration into the host OS.
This makes it invisible to attackers and avoids slowing down the system it protects. It works across containerized and multi-tenant environments, including deployments using Nvidia NIM microservices.
Nvidia says Argus uses memory forensics to detect threats up to 1,000 times faster than existing agentless solutions. Because it doesn’t run on the host, it won’t impact performance.
The system is built to work with existing enterprise security setups like SIEM, SOAR, and XDR platforms.
Cisco is working with Nvidia to create a Secure AI Factory using this architecture.
“Now is the time for enterprises to be driving forward with AI, but the key to unlocking innovative use cases and enabling broad adoption is safety and security,” said Jeetu Patel, executive vice president and chief product officer at Cisco.
DOCA Argus is part of Nvidia’s broader cybersecurity AI platform, which includes BlueField hardware and the Morpheus AI framework. The goal is to improve visibility and threat response across AI infrastructure.
Argus is trained using Nvidia’s internal security data, with the aim of reducing false alerts and focusing only on real threats. This, the company says, will help security teams avoid alert fatigue and act faster
