Zayo and NVIDIA announce effort to expand network capacity for AI “factories”

Zayo, a digital infrastructure company based in Denver, said it has teamed with NVIDIA to expand “critical network capacity” aimed at powering what the companies described as today’s AI ecosystem and the next generation of AI innovation. The announcement, published Aug. 6, 2026, positions Zayo as a connectivity supplier to the growing stack of companies building and operating data-center capacity for AI training and inference.

In the announcement shared via Yahoo Finance, the companies framed the work around scaling network resources that can carry the data flows associated with large AI systems. The term “AI factories” is commonly used in the industry to describe clusters of data centers and associated infrastructure built specifically to run AI workloads at scale, typically requiring significant throughput and dependable network performance.

While Zayo did not provide operational specifics in the available account, it characterized the effort as “critical network capacity” delivery tied to NVIDIA’s AI technology ecosystem. Partnerships at this layer often focus on ensuring that bandwidth, routing, and interconnection options can support the movement of large model artifacts, training data, and inference traffic between data-center sites, cloud regions, and technology partners.

The announcement also reflects NVIDIA’s broader role in accelerating AI infrastructure demand. NVIDIA’s data-center platform is widely used for AI computation, and that computation increasingly depends on network systems that can keep GPUs fed and interconnect systems efficiently. In this context, connectivity providers like Zayo are drawn into the picture because AI capability is not limited to chips and servers; it also relies on how quickly and reliably data can move across the stack.

From Zayo’s standpoint, the statement underscores a demand announcement it wants to capture as customers expand data-center and interconnection footprints for AI. Network providers typically compete on reach, latency, and the ability to scale capacity as demand rises, especially for customers that require repeatable performance across multiple sites. Even without disclosed contract terms, the language suggests Zayo is aligning its capacity planning with the infrastructure build-out that AI customers are pursuing.

As of the Aug. 6 announcement, the companies did not disclose the size of the investment, the geographic scope, the timeline, or the specific network components involved. They also did not provide information on whether the relationship is structured as a multi-year capacity agreement, a network services bundle, or a targeted deployment supporting a particular NVIDIA platform or customer deployment. Those details would be necessary to assess how material the partnership may be to Zayo’s revenue mix or capital spending profile.

For investors and customers tracking AI infrastructure build-outs, the near-term question is whether Zayo and NVIDIA will offer additional operational details in subsequent releases, such as deployment locations, capacity milestones, or performance targets. Watch for follow-on communications that clarify the commercial structure and how the network expansion maps onto specific AI data-center plans and interconnection requirements. Until then, the announcement is best read as a high-level announcement that connectivity and AI compute roadmaps are being planned together.