Nvidia’s latest chip designs are now in customers’ hands, and with an estimated 80-81% share of the data-center AI GPU market according to IDC data, the company isn’t just leading the AI chip race — it’s lapping the field.

The company confirmed that its newest hardware is shipping to customers. Bitcoin miners are increasingly repurposing their operations to serve AI workloads, creating a new intersection between the GPU supply chain and the digital asset industry.

The Rubin platform and what’s actually shipping

The Rubin platform, Nvidia’s next-generation architecture, entered full production in early 2026. Partner systems built on Rubin are expected to become available in the second half of this year. The first cloud deployments of Rubin-based instances are anticipated from AWS, Google Cloud, and Microsoft.

H200 AI chips have begun shipping to China, though in minimal volumes as of mid-July. US export restrictions continue to limit what Nvidia can sell into the Chinese market.

Bitcoin miners pivot to AI, and Nvidia benefits either way

Bitcoin miners, sitting on massive GPU infrastructure and access to cheap power, are increasingly redirecting their resources toward AI workloads. This shift means that instead of using Nvidia GPUs exclusively for traditional crypto mining operations, these companies are repurposing their hardware and facilities to serve the AI industry.

What this means for investors

Nvidia’s 80-81% market share in data-center AI GPUs means that virtually every major AI deployment flows through Nvidia hardware. Nvidia’s CUDA software ecosystem creates substantial lock-in effects for developers and enterprises, making its offerings indispensable for training and inference in nearly all major AI labs and cloud service providers.

For crypto-adjacent investors, the Bitcoin miner pivot toward AI represents a potential valuation catalyst for publicly traded mining companies. Firms that can demonstrate meaningful AI hosting revenue alongside their mining operations may command higher multiples, since AI revenue is viewed as more predictable and less correlated to volatile crypto prices.

AMD, Intel, and a growing number of custom silicon efforts from cloud providers are all competing for Nvidia’s position. However, an 80-81% market share doesn’t erode overnight, especially given the high switching costs in enterprise compute tied to Nvidia’s architecture.