Cisco and NVIDIA have announced an expanded partnership focused on AI infrastructure. As part of the collaboration, Cisco’s Silicon One chip will be integrated into NVIDIA’s Spectrum-X platform.
A recent Cisco survey of business leaders found that 97 percent of respondents view AI projects positively, while only 1.7 percent feel fully prepared for the transition. The same survey identified three main obstacles slowing adoption: a pressing need to modernize infrastructure, concerns about secure and responsible AI use from a cybersecurity perspective, and a shortage of the necessary skills to implement and maximize AI investments.
”Until now, most AI conversations have focused on chips, compute power and large language models. As organizations begin to run AI workloads, however, the real challenge will be connecting compute resources and data — both within and between data centers and the cloud — and it’s this connectivity that will define the limits of AI innovation. The network becomes essential to scaling AI across the enterprise,” says Jeetu Patel, EVP and Chief Product Officer at Cisco.
Cisco and NVIDIA are working together to ease these transitions and are deepening their partnership with several expanded and new initiatives for jointly developing networking solutions that meet the demands of AI transformation.
Among other things, the collaboration means NVIDIA will enable Cisco Silicon One to be part of the NVIDIA Spectrum-X Ethernet networking platform. With this move, Cisco becomes the only partner whose chip is included in NVIDIA Spectrum-X.
”This is not only huge validation for Silicon One, it also highlights how central Cisco technology — including chip design, optics and software — is to the AI ecosystem,” says Jeetu Patel.
Cisco will also build systems that combine NVIDIA Spectrum chips with Cisco OS software, allowing customers to standardize on both Cisco networking and NVIDIA technology in the data center. That combination aims to support a broad range of AI use cases and applications while simplifying deployment and operations for enterprises.
The expanded partnership addresses the three challenges identified by business leaders: by integrating advanced networking hardware and software, the companies aim to modernize infrastructure; by focusing on security and interoperability they seek to support responsible AI practices; and by delivering integrated solutions they hope to reduce the skills gap and accelerate enterprise adoption of AI.
As organizations scale AI beyond experimentation, the emphasis will shift from isolated compute capabilities to a holistic architecture where high-performance networking, storage, and software work together. Cisco and NVIDIA’s joint efforts are positioned to make it easier for enterprises to deploy scalable, secure, and operationally efficient AI systems across data centers and cloud environments.
By combining Cisco’s networking expertise and software stack with NVIDIA’s GPU and networking innovations, the partnership aims to streamline the path from pilot projects to production AI deployments, helping organizations extract greater value from their data and compute investments while maintaining control over security, compliance and operational complexity.