This week, AI industry heavyweights Cisco and NVIDIA are expanding a joint partnership to accelerate enterprise AI projects through each firm’s extensive data centre networks.
Following a pre-existing Cisco/NVIDIA collaboration to optimize Cisco Nexus Hyperfabric with AI features, Cisco’s Silicon One product is now coupled with NVIDIA SuperNICs and joining the NVIDIA Spectrum-X Ethernet networking platform.
This marks NVIDIA’s first and currently only partner silicon solution integrated into its Spectrum-X network. Following the Spectrum-X integration, Cisco is also working on a framework combining NVIDIA Spectrum silicon with Cisco’s first-party OS software for enhanced back-end connectivity industry interoperability.
Cisco has big plans for the NVIDIA partnership in the future. Together, the firms look to grow investments in cross-portfolio innovations. This move will help Cisco answer common hurdles such as congestion management and load balancing, allowing for accelerated AI deployments.
Cisco Launches AI Security Solutions at Recent AI Summit
Cisco announced two major security products at last month’s AI Summit. They aim to tackle the security challenges associated with the rapid infrastructure expansion. The launch of AI Defense and Hypershield comes when enterprises are increasingly concerned about securing AI operations at scale.
Scheduled for release in March 2025, AI Defense targets two critical security gaps in enterprise AI adoption. For developers, it offers automated testing of AI models, checking for hundreds of potential safety and security vulnerabilities while providing continuous runtime validation against threats such as prompt injection and denial-of-service attacks.
Security teams will benefit from enhanced visibility into AI tool usage across the organization. This will allow them to control employee access to AI applications and prevent sensitive data leakage. This solution integrates with Cisco’s Security Cloud platform and utilizes network-level security infrastructure to improve threat detection.
Kent Noyes, Global Head of AI and Cyber Innovation at World Wide Technology, remarked:
The adoption of AI exposes companies to new risks that traditional cybersecurity solutions don’t address. Cisco AI Defense represents a significant leap forward in AI security, providing full visibility of an enterprise’s AI assets and protection against evolving threats.
Hypershield represents a significant advancement in data centre security architecture. Its AI-driven rule engine can identify and address vulnerabilities in minutes rather than months by deploying precise security controls tested against live production traffic.
The system features self-adapting network segmentation that evolves continuously based on process behaviours and file changes. Hypershield validates security updates and policy changes against live traffic without compromising performance. Additionally, its architecture seamlessly incorporates new security enforcement points, ensuring scalability as organizations expand their AI infrastructure.
Tom Gillis, SVP and General Manager at Security Business Group, discusses Hypershield’s capabilities in the context of the scale of data centres needed to run enterprise-level AI systems:
AI is transformative, driving huge productivity gains. The engine of AI — the data centre — will grow substantially, maybe an order of magnitude or more over the coming years.
Every enterprise must consider how to transition to an “AI-scale data centre,” as traditional appliances will not suffice for this shift. I am proud to introduce Cisco Hypershield, the first truly distributed, AI-native system that integrates security directly into every software component of all applications running on your network—across every server and in public and private cloud deployments.
Only 13% of customers are fully prepared for AI deployment, while 90% want to harness its benefits within the next two years. These solutions aim to address urgent security needs. Early adopters in sectors like healthcare, manufacturing, and service industries showcase AI’s potential while revealing significant security challenges.
