Nvidia exits public cloud business to refocus on AI chip R&D and platform software layers.

Nvidia has reorganized its cloud computing group, folding the DGX Cloud business into its core engineering structure under the leadership of SVP Dwight Diercks, who oversees software engineering. According to reporting by The Information, this restructuring represents a significant scaling back of Nvidia's ambitions to operate a public cloud service that would directly compete with Amazon Web Services. Rather than continuing to sell GPU compute as a service under its own brand, the company is repositioning DGX Cloud as an internal platform for its engineers, with a focus on chip demand and AI model development.

DGX Cloud was introduced in early 2023 as Nvidia's attempt to abstract its flagship DGX systems into a managed service, hosted initially on infrastructure provided by AWS, Google Cloud, Oracle Cloud, and Microsoft Azure. The service offered dedicated H100-based clusters with Nvidia's full software stack preinstalled, providing an attractive pitch to enterprise customers who could rent Nvidia's preferred AI platform without building their own data centers. In practice, however, the model proved difficult to scale. Pricing was high compared to commodity GPU instances, integration with existing cloud tooling was uneven, and support responsibilities were split between Nvidia and its hosting partners. Customers running DGX Cloud across multiple providers faced operational complexity, while hyperscalers were rapidly cutting prices on H100 capacity and rolling out their own managed AI services.

The restructuring makes strategic sense given that Nvidia's largest customers are the very companies it would have been competing against. AWS, Microsoft, Google, and other cloud providers account for a significant share of Nvidia's data center revenue, and running a first-party cloud service risked creating channel conflict at a time when those customers are committing billions of dollars to Nvidia hardware. Operating a competitive cloud platform also requires sustained capital expenditure on facilities, networking, power, and operations—advantages that lie with the hyperscalers, not with Nvidia. Additionally, hyperscalers have strong incentives to differentiate above Nvidia's hardware layer: AWS invests in Trainium and Inferentia accelerators, Google pushes TPUs, and Microsoft is expanding its Maia program, which recently partnered with Intel Foundry to build chips.

There were also signs of strain in GPU supply dynamics. Nvidia has increasingly acted as both supplier and customer, leasing back large volumes of GPU capacity from cloud providers and specialized operators—in one case, 18,000 GPUs to the tune of $1.5 billion over four years. This arrangement becomes harder to justify if Nvidia is simultaneously selling competing cloud services. By stepping back and restructuring DGX under its engineering business, Nvidia simplifies those relationships, repositioning DGX Cloud as a tool that helps partners deploy Nvidia hardware more effectively rather than as a product that competes with their own offerings. The reorganization does not necessarily mean reduced investment in cloud-adjacent technologies. Nvidia continues to expand its software stack across inference, orchestration, networking, and systems management, with platforms such as CUDA and TensorRT designed to run everywhere hyperscalers operate.