NVIDIA DOCA Software Framework Offers Agentless Runtime Threat Detection with Zero Overhead

NVIDIA is debuting a new software framework for securing AI factories—or where complex, agentic workflows are executed, as NVIDIA defines. This framework, dubbed NVIDIA DOCA, brings greater security power to the NVIDIA BlueField networking platform, enabling the immediate detection of and response to attacks on AI workloads.

The NVIDIA DOCA Argus framework, through the power of advanced memory forensics, delivers runtime threat detection that accelerates detection by up to 1,000x compared to other agentless solutions. Running independently of the host and without impacting system performance, Argus forwards an agentless, zero-overhead design that increases system efficiency while persisting resilient security. And, by operating outside of the host, Argus remains invisible to attackers, even during a system compromise, according to NVIDIA.

“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.”

Argus integrates seamlessly into enterprises’ SIEM, SOAR, and XDR security platforms, helping extend existing security investments. The combination of NVIDIA BlueField’s built-in, scalable, data-centric protection for AI workloads with DOCA Argus’ proactive threat detection enables organizations to effectively secure their AI factories without making sacrifices to performance or efficiency, NVIDIA noted. Additionally, Argus is purpose-built to protect containerized workloads—such as NVIDIA NIM microservices—integrating security at every level of the AI application stack.

As part of this announcement, NVIDIA and Cisco are collaborating to develop a Secure AI Factory with NVIDIA architecture that streamlines AI deployment and management at scale.

“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. “NVIDIA and Cisco are providing enterprises with the infrastructure they need to confidently scale AI while safeguarding their most valuable data.”

To learn more about NVIDIA DOCA, please visit https://www.nvidia.com/en-us/.