Cisco and NVIDIA Revolutionize AI Deployment with Secure AI Factory Expansion

Transforming AI Deployment: Cisco and NVIDIA's Secure AI Factory

In a significant announcement on March 16, 2026, Cisco expanded its Secure AI Factory, developed in partnership with NVIDIA, to enable organizations to seamlessly deploy AI across a broader spectrum of environments. This enhancement addresses the pressing need for businesses to implement AI efficiently and securely across data centers and edge sites where quick decision-making is crucial.

Revolutionizing Infrastructure Flexibility

The newly expanded architecture allows enterprises to run AI applications at scale, from massive central data centers all the way down to local environments like hospitals, warehouses, and even moving vehicles. According to Chuck Robbins, Cisco's Chair and CEO, many organizations recognize AI's potential to transform operations, but they struggle to deploy it safely at scale. This expansion aims to eliminate those hurdles by providing a cohesive framework for integrating AI into existing systems rather than forcing companies to stitch together disparate technologies.

Robbins emphasized that Cisco's collaboration with NVIDIA aims to raise performance standards while simplifying the deployment and security of AI infrastructure. This is vital as businesses move from pilot projects to large-scale production while ensuring security is baked into the system from the very beginning.

Edge AI Inference and Efficiency

AI is no longer confined to centralized data centers. The need for real-time decision-making necessitates that AI inference occurs closer to the data sources. For example, AI applications in healthcare may need to analyze patient data instantaneously, while manufacturing might require real-time video analysis to keep factory workers safe. Cisco and NVIDIA are empowering enterprises to address these edge inferencing use cases effectively.

To achieve this, Cisco has commenced support for NVIDIA RTX PRO™ 4500 Blackwell Server Edition GPUs across its Unified Computing System (UCS) and Unified Edge portfolios. This transformation allows organizations to run mission-critical AI workloads at the edge, significantly reducing both energy consumption and infrastructure costs typically associated with large data center setups.

Furthermore, the Cisco AI Grid offers a reference design that combines Cisco’s Mobility Services Platform with NVIDIA GPUs, enabling service providers to deliver edge AI applications effectively.

Enhancements in Performance for Large-Scale AI Deployments

Cisco's innovations don’t stop at edge computing. With the recent introduction of high-speed switches, including the astounding 102.4Tbps Cisco N9100 powered by NVIDIA Spectrum-6 Ethernet switch silicon, the company aims to push the boundaries of AI performance even higher. Organizations can now rely on Cisco Nexus Hyperfabric, integrated into Cisco Nexus One, to transform complex, multi-vendor systems into cohesive, full-stack solutions, drastically reducing deployment times and easing IT burdens.

Customers will also appreciate the ability to select from validated architectures, making it readily adaptable for a myriad of AI factory implementations, each complying with NVIDIA Cloud Partner program specifications.

Security at Every Layer

As the landscape of AI continues to evolve and models become high-value assets subject to exploitation, Cisco is intensifying its security measures. By embedding security protocols deeply across the infrastructure, Cisco ensures that both external threats and rogue AI agent behaviors are addressed.

For instance, the Hybrid Mesh Firewall ensures that consistent security policies are applied across multiple enforcement points. This includes extending policy enforcement capabilities to NVIDIA BlueField data processing units (DPUs), blocking threats before they infiltrate organizational networks. This innovative approach guarantees that AI workloads maintain robust security without sacrificing performance.

Additionally, Cisco AI Defense will introduce critical security features for AI agents at the edge, integrating with NVIDIA's offerings to provide automated vulnerability testing and necessary guardrails. This integration allows AI developers to remain proactive against emerging threats, reinforcing users' trust in their AI implementations.

Bridging Innovation with Security

With the modern AI landscape increasingly populated by autonomous agents, the responsibility for security cannot simply be an afterthought. Cisco's commitment to safeguarding enterprise AI deployment resonates through its innovative security offerings. The introduction of controls for managing AI agents, including those running on the NVIDIA OpenShell, underscores Cisco's vision of empowering organizations to harness AI without compromising their security posture.

Industry leaders are already expressing their excitement over these developments. Cirrascale's CTO, Alex Nataros, praised the advancements in Cisco's high-performance networking solutions, which enhance cloud flexibility while ensuring operational simplicity. Similarly, Sharon AI's COO, Andrew Leece, looked forward to utilizing Cisco's infrastructure for scaling AI with substantial performance.

Conclusion

As organizations evolve in their approaches to adopting AI, Cisco's expanded Secure AI Factory with NVIDIA presents a robust framework to meet the dual challenges of efficiency and security. By equipping businesses with the necessary tools to deploy AI seamlessly across their infrastructures, this collaboration not only enhances operational capabilities but also builds a future where organizations can trust their increasingly autonomous systems to function securely and reliably.