Nvidia to Invest $2.1 Billion in IREN to Scale AI Data Center Capacity

In a move that further cements its role as the primary architect of the global artificial intelligence boom, Nvidia has reached an agreement to invest up to $2.1 billion in data center operator IREN. The deal, which aims to accelerate the deployment of large-scale AI infrastructure, comes as the world’s most valuable chipmaker seeks to secure the physical capacity required to house its next generation of high-performance hardware.

According to www.reuters.com, the partnership is part of a broader strategic initiative to deploy up to 5 gigawatts (GW) of power capacity. This massive scale is intended to meet the unrelenting demand for AI training and inference from frontier model developers and hyperscale cloud providers.

The Strategic Shift Toward ‘Neocloud’ Infrastructure

The investment highlights the growing importance of “neocloud” providers—specialized firms that build cloud computing environments specifically optimized for Nvidia’s GPU architectures. Unlike traditional hyperscalers that develop their own custom silicon, neoclouds offer a direct pipeline for enterprise customers and Big Tech firms to access massive clusters of Nvidia H100, H200, and Blackwell-class processors without the overhead of building their own physical facilities.

As reported by www.reuters.com, IREN has granted Nvidia a five-year right to purchase up to 30 million shares at an exercise price of $70 per share. This equity-linked structure aligns the interests of the hardware manufacturer with the infrastructure operator, ensuring that the physical sites are designed according to Nvidia’s specific “AI factory” blueprints.

For those looking to compare providers in the increasingly crowded market for high-end compute, this deal signals that IREN is positioning itself as a top-tier destination for enterprise-grade AI workloads.

Texas Sweetwater Campus: A 2-Gigawatt Hub

A primary focus of the future deployment will be IREN’s Sweetwater campus in Texas. The site is slated to reach 2 gigawatts of capacity, making it one of the largest dedicated AI data center complexes in the world. The availability of power and high-density cooling infrastructure has become the primary bottleneck for AI expansion, and the Sweetwater facility is designed to address these constraints directly.

The partnership aims to combine Nvidia’s reference architecture for AI factories with IREN’s operational expertise in managing high-power-density environments. This vertical integration is becoming a necessity as the power requirements for modern GPU specifications continue to climb, often exceeding the capabilities of older, legacy data centers.

Market Context and Big Tech Spending

The timing of the deal coincides with a period of unprecedented capital expenditure from the world’s largest technology companies. Microsoft, Alphabet, Meta, and Amazon recently signaled that their combined AI-related spending is set to surpass $700 billion this year. This deluge of capital is flowing directly into GPU cloud providers and infrastructure firms like IREN that can provide the immediate capacity these giants require.

IREN is no stranger to large-scale enterprise contracts. Last year, the company secured a $9.7 billion cloud deal with Microsoft, establishing its credibility as a partner for hyperscale-level operations. The new Nvidia investment further validates IREN’s pivot toward high-performance computing (HPC) and AI-centric services.

Industry Implications for AI Compute

Nvidia’s decision to invest directly in an infrastructure partner suggests a shift in strategy. By financing the build-out of 5GW of capacity, Nvidia is essentially ensuring that there is a “home” for the millions of GPUs it plans to ship over the next several years. This move mitigates the risk of supply chain bottlenecks shifting from chip manufacturing to data center availability.

Industry analysts view this as a defensive and offensive maneuver. Defensively, it secures the power and space needed to maintain sales momentum. Offensively, it allows Nvidia to influence the design of the data centers themselves, ensuring they are optimized for the liquid cooling and high-speed networking required by their latest Blackwell platforms.

Following the announcement, shares of IREN rose approximately 9% in extended trading, reflecting investor confidence in the company’s expanded role within the AI ecosystem. As the race for compute continues, the alliance between the world’s leading AI chipmaker and a major infrastructure operator marks a significant milestone in the industrialization of artificial intelligence.

For more details on the financial specifics of the agreement, visit the official report at www.reuters.com.