Nvidia has committed roughly $2 billion to AI cloud infrastructure company Nebius through pre-funded warrants, taking a minority stake while helping finance a major expansion of hyperscale computing capacity. The deal underscores a broader industry shift: artificial intelligence is no longer just a software story. It is increasingly a battle over electricity, advanced chips, cooling systems, and the ability to build enormous data center campuses fast enough to meet demand.
According to the announcement, Nebius aims to build more than 5 gigawatts of Nvidia-accelerated computing capacity by 2030. That scale is far beyond the footprint of a typical cloud expansion plan. The infrastructure is intended to support large-model training, inference workloads, and the growing wave of agentic AI systems now driving a new surge in compute demand.
A Strategic Bet on AI-Native Cloud Infrastructure
Nebius, headquartered in Amsterdam and listed on Nasdaq under the ticker NBIS, is part of a new generation of cloud providers built specifically for AI workloads. Unlike legacy cloud platforms that adapted existing enterprise infrastructure for AI use, Nebius is positioning itself as an AI-native operator designed around dense GPU clusters and modern accelerator-driven computing.
The company’s origins trace back to the restructuring of Yandex’s international business. In 2024, after the Dutch parent of Russian search giant Yandex divested its Russian operations, the remaining international assets were reorganized into Nebius. Since then, the company has expanded AI infrastructure operations across the United States, Europe, and Israel.
Nebius already runs large GPU clusters powered by Nvidia hardware, including H100 and H200 accelerators. Under the new partnership, it also plans to adopt future Nvidia technologies such as the Rubin architecture, Vera CPUs, and BlueField storage platforms. The collaboration spans the full stack, from data center design to inference deployment and fleet-level GPU management.
For Nvidia, the logic is straightforward. By backing Nebius, the chipmaker is not only investing in a fast-growing infrastructure company but also securing a long-term ecosystem partner likely to consume vast quantities of Nvidia silicon. That makes the transaction strategically important well beyond its financial size.
From AI Software Boom to Industrial-Scale Buildout
Nvidia CEO Jensen Huang described the partnership as part of the next stage of the AI cycle, pointing to the rise of agentic AI and the resulting jump in infrastructure demand. Nebius CEO Arkady Volozh similarly emphasized that the company was built specifically for AI from the start, rather than retrofitted from older computing models.
The message from both companies is clear: the next phase of AI growth depends on physical buildout. As enterprises and developers push larger models into production and inference demand scales globally, the underlying constraints are becoming more obvious. Access to GPUs matters, but so do land, power contracts, cooling design, and the ability to deploy at industrial speed.
That reality is increasingly reshaping the economics of AI. What was once framed primarily as a race among model developers is now also a contest among infrastructure providers capable of turning chip supply into usable, power-intensive computing capacity.
Missouri Project Highlights the Scale of Ambition
Just days before the investment was announced, Nebius secured approval for a 1.2-gigawatt AI data center campus in Independence, Missouri. The project is described as one of the largest planned AI infrastructure developments in the United States.
Local projections suggest the site could generate around 1,200 construction jobs and produce an estimated $650 million in economic impact over a 20-year period. While those figures reflect just one location, they offer a glimpse into the scale at which AI infrastructure is now being planned.
The Missouri campus is only one component of a much larger expansion strategy. Nebius said it could operate up to 16 global data center sites by the end of 2026, with contracted power capacity approaching 3 gigawatts within the next year. If executed on schedule, that would place the company among the more aggressive builders in the fast-emerging AI infrastructure segment.
Heavy Spending, Modest Revenue, Big Contract Pipeline
Nebius is spending at a pace that reflects the capital intensity of the market it is chasing. The company reported about $2.1 billion in capital expenditures in the fourth quarter of 2025 alone. For 2026, projected spending is expected to range between $16 billion and $20 billion.
By comparison, revenue remains relatively small. Nebius reported roughly $530 million in trailing 12-month revenue. On a traditional valuation basis, that gap between present revenue and planned investment is striking. But the company argues that long-term contracts and backlog support a much larger future revenue base.
Among the notable agreements cited are a multi-year deal with Microsoft tied to a data center project in Vineland, New Jersey, estimated at between $17 billion and $19.4 billion, as well as a separate $3 billion arrangement with Meta Platforms. These agreements suggest that major technology buyers are already looking beyond chip procurement and securing capacity from specialized AI infrastructure operators.
Market Reaction and the “Nvidia Halo Effect”
Investors responded positively to the news. Nebius shares rose roughly 13% to 16% following the investment announcement, lifting the company’s market capitalization above $24 billion in early trading. The move reflects a pattern frequently seen in AI markets: when Nvidia commits capital or technology to a partner, investors often treat it as a strong endorsement of that company’s position in the ecosystem.
This so-called “Nvidia halo effect” has become an important force in the public market narrative around AI infrastructure. Nvidia is not merely the dominant supplier of accelerators; it is also increasingly acting as an ecosystem architect, helping shape where capacity gets built and which providers are likely to emerge as major platforms for the next generation of AI deployment.
Why This Deal Matters Beyond One Company
The Nvidia-Nebius partnership highlights a central truth now emerging across the AI industry: algorithms may capture headlines, but infrastructure determines scale. As more advanced models move from research into commercial use, the bottlenecks are shifting toward the physical layer—power availability, cooling, networking, advanced chips, and large-scale data center construction.
In that context, Nvidia’s investment in Nebius is not just a minority stake in a cloud company. It is a signal that the race to build the backbone of AI is accelerating, and that the winners may be those capable of combining software demand with industrial execution. If Nebius can deliver on its planned gigawatt-scale footprint, it could become one of the more significant AI-native infrastructure players in the market. If not, the announcement still makes one thing unmistakable: the future of AI will be built not only in code, but in concrete, copper, and silicon.