Market Analysis
Data Centers Are the New Skyscrapers and AI Is the Reason
Rahul Bablani•
One of the biggest trends happening in the economy right now is something most people do not see every day. Across the United States and around the world, companies are building data centers at an incredible pace. In fact, construction spending on data centers is now catching up to and even overtaking office building construction, which is a huge shift in what companies are investing in.
A big reason for this boom is the explosion of artificial intelligence. AI systems require massive amounts of computing power, and all of that computing happens inside large facilities filled with servers called data centers. As AI continues to grow, the demand for these facilities is growing even faster.
Because of that, a new category of companies is starting to get a lot of attention from investors: AI infrastructure and data center stocks.
Data Centers Are Replacing Office Buildings
For decades, office towers were one of the most important types of commercial real estate. Cities spent billions of dollars building offices for companies and workers.
But that trend has been changing since the pandemic. Many companies have shifted to remote or hybrid work, which means they need less office space than before.
At the same time, demand for data centers has exploded.
Recent construction data shows that office construction spending has fallen significantly, while spending on data centers has surged dramatically over the past decade.
A little over ten years ago, data center construction was almost negligible compared to office construction. Now it is approaching tens of billions of dollars per year and continuing to grow quickly.
This shift shows how the economy itself is changing. Instead of building places for people to work physically, companies are building places for computers to work digitally.
AI Is Driving the Data Center Boom
The biggest driver behind this construction boom is artificial intelligence.
Training and running AI models requires an enormous amount of computing power. Large AI models need thousands of GPUs working together, often running continuously for weeks or months.
Those GPUs are housed inside data centers that consume huge amounts of electricity and cooling systems.
In fact, analysts expect global electricity demand from data centers to double in the coming years, largely because of AI workloads.
Tech companies like Meta, Microsoft, and Google are racing to secure as much computing capacity as possible because whoever controls the infrastructure will have a major advantage in the AI race.
That is why data center companies are suddenly becoming extremely important.
Nebius Just Made a Massive Deal With META
One of the biggest recent developments in this space involves Nebius Group, which trades under the ticker NBIS.
Nebius focuses on building AI cloud infrastructure and data centers that run high performance GPUs used for training artificial intelligence models.
Today it was reported that Nebius signed a massive agreement with Meta worth up to $27 billion to provide AI computing capacity over the next several years.
Under the deal, Meta will purchase about $12 billion of AI computing capacity initially, with the possibility of expanding the agreement to $27 billion over five years.
This is a huge deal because it shows just how much demand there is for AI infrastructure right now. Tech companies are willing to commit billions of dollars just to make sure they have enough computing power to run their AI models.
Nebius also recently received a $2 billion investment from NVIDIA, which also shows how important the company’s data center infrastructure could become in the AI ecosystem.
Because of this news, NBIS has become one of the most talked about AI infrastructure stocks in the market.
Other AI Data Center Stocks Investors Are Watching
While Nebius Group has been getting a lot of attention because of its massive new deal with Meta Platforms, it is not the only company investors are watching in the AI data center space. There are a few other companies that are becoming increasingly important because they provide the computing infrastructure that powers artificial intelligence.
The reason these companies are interesting is because they are basically building the physical backbone of the AI economy. AI models cannot run without huge amounts of computing power, and that computing power requires massive data centers filled with GPUs and advanced networking equipment.
Two companies that are often mentioned alongside Nebius are IREN Limited and CoreWeave
IREN: Renewable-Powered AI Data Centers
IREN is an interesting company because it originally started in the digital infrastructure space supporting crypto mining operations. However, over the past few years the company has been shifting its focus toward AI data center infrastructure.
Today IREN builds and operates high-performance computing data centers that are designed to support AI workloads like model training and inference. One of the things that makes the company unique is that many of its facilities are powered by renewable energy sources, which helps address one of the biggest challenges facing the AI industry: electricity consumption.
AI training clusters require enormous amounts of power, and some analysts believe electricity availability could become one of the biggest bottlenecks for AI growth in the future. Because IREN focuses on renewable-powered data centers and efficient energy infrastructure, some investors see it as a company that could benefit from this demand.
The company has also signed major infrastructure agreements with large technology companies. For example, IREN previously secured a $9.7 billion agreement with Microsoft to provide GPU cloud infrastructure for AI workloads, which highlights how large tech companies are increasingly relying on specialized data center providers.
As AI demand continues to grow, companies like IREN that already operate large-scale computing facilities could play a much bigger role in supplying the infrastructure needed to train future AI models.
CoreWeave: The “GPU Cloud” Company
Another major player in this space is CoreWeave, which trades under the ticker CRWV. CoreWeave focuses specifically on providing GPU cloud services designed for AI companies and developers.
Instead of building AI models itself, CoreWeave rents out access to powerful GPUs that companies use to train and run artificial intelligence systems. This model is sometimes described as GPU-as-a-Service, meaning companies can rent computing power instead of building their own data centers.
CoreWeave has grown extremely quickly because demand for GPUs has exploded with the rise of generative AI. The company operates large clusters of GPUs across multiple data centers and has become a key infrastructure provider for AI developers.
One of the reasons CoreWeave has been successful is its strong relationship with NVIDIA, whose GPUs are widely used for AI training. NVIDIA has even taken a strategic stake in the company, which signals confidence in CoreWeave’s role in the AI ecosystem.
Because AI companies need access to thousands of GPUs at once, platforms like CoreWeave allow them to scale quickly without building their own infrastructure.
Why These Stocks Are Being Called “Picks and Shovels”
Many investors describe companies like Nebius, IREN, and CoreWeave as “picks and shovels” plays for AI.
The phrase comes from the California gold rush. During the gold rush, the people who consistently made money were not always the miners searching for gold. Instead, the companies selling tools like picks and shovels often did very well because everyone needed their equipment.
In the AI boom, companies building data centers and GPU infrastructure play a similar role. Every AI startup, tech company, and research lab needs computing power.
That means the companies supplying that infrastructure could benefit regardless of which AI models or software platforms end up dominating the market.
The Bigger Picture
What is happening right now with data centers shows how much the economy is shifting toward digital infrastructure.
In the past, companies invested heavily in physical infrastructure like office buildings, factories, and retail space. Today, many of the most valuable assets in the economy are data centers, cloud infrastructure, and computing power.
Artificial intelligence is accelerating this trend even further. Every AI model, chatbot, image generator, and recommendation algorithm needs enormous computing resources behind the scenes.
That means the companies building the infrastructure that powers AI could end up being some of the biggest long term winners.
The Bigger Picture
What is happening right now with data centers shows how much the economy is shifting toward digital infrastructure.
In the past, companies invested heavily in physical infrastructure like office buildings, factories, and retail space. Today, many of the most valuable assets in the economy are data centers, cloud infrastructure, and computing power.
Artificial intelligence is accelerating this trend even further. Every AI model, chatbot, image generator, and recommendation algorithm needs enormous computing resources behind the scenes.
That means the companies building the infrastructure that powers AI could end up being some of the biggest long term winners.