NVIDIA Launches DOCA Argus for Real-Time, Agentless AI Security
As businesses increasingly leverage artificial intelligence (AI), securing AI factories, spaces where complicated, agentic workflows are enacted has become more important than ever.
NVIDIA is filling this requirement by bringing to market the NVIDIA DOCA Argus software framework, an integral element of the NVIDIA cybersecurity AI platform. This new product delivers runtime AI workload security, running flawlessly on the NVIDIA BlueField network platform, identifying and mitigating attacks in real time.
Agentless Threat Detection at Unparalleled Speed
The DOCA Argus solution employs cutting-edge memory forensics to provide runtime threat detection at speeds of up to 1,000x that of conventional agentless solutions.
Crucially, this is done without affecting system performance, meaning security is strong but not at the expense of efficiency. Argus is unlike traditional tools in that it is host-independent, meaning no agents or host-based resource integration is required, enhancing system efficiency.
This zero-overhead, agentless architecture also guarantees Argus's transparency to attackers in case the system is breached, making it particularly suitable for AI compute environments, such as multi-tenant and containerized infrastructure.
Seamless Integration with Existing Security Platforms
Cybersecurity experts can seamlessly integrate DOCA Argus with current security systems like SIEM, SOAR, and XDR to enable real-time monitoring, automated threat reduction, and improved security on AI infrastructure. This integration enables the extension of the functionality of existing cybersecurity systems, keeping AI factories secure with minimal interference with ongoing operations.
Built for Scalable AI Security with NVIDIA BlueField
NVIDIA BlueField is the foundational security component for every AI factory, offering data-centric protection at scale for AI workloads. By combining BlueField’s acceleration capabilities with DOCA Argus’ proactive threat detection, enterprises can secure their AI factories without sacrificing performance or efficiency.
Cisco is partnering with NVIDIA to deliver a Secure AI Factory using NVIDIA’s architecture. This collaborative effort simplifies the deployment and protection of AI infrastructure at scale by embedding security at every layer of the AI factory from the outset, rather than adding it as an afterthought.
Also read: Amazon and Nvidia Signal Fossil Fuels May Play Role in Powering AI Growth
Jeetu Patel, executive vice president and chief product officer at Cisco, said, “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.”
He added, “NVIDIA and Cisco are providing enterprises with the infrastructure they need to confidently scale AI while safeguarding their most valuable data.”
As AI adoption and data volumes grow, identifying real threats becomes harder—especially with the rise of agentic AI. NVIDIA’s DOCA Argus addresses this by minimizing false positives and focusing on validated threats.
Designed for containerized workloads like NVIDIA NIM microservices, it uses real-world threat intelligence to secure every layer of the AI stack.
