What Are AI Data Center Stocks?
If you have been following the market over the past two years, you have probably heard investors talk about the “AI trade.” Most people immediately think of companies like NVIDIA or ChatGPT.
But there is another part of the AI story that many investors still do not fully understand.
AI data center stocks are the companies helping build the physical infrastructure needed to make artificial intelligence work. These are the businesses supplying the cooling systems, electricity, fiber optics, construction equipment, and electrical components that power AI data centers.
Think about it this way. Every time AI models get smarter, they require more computing power. More computing power means more servers. More servers mean more data centers.
And data centers need to be built.
That is where this opportunity lies.
No, We Do Not Think It Is Too Late
Many of the stocks tied to the AI data center buildout have already had major runs. Some have doubled. Others have moved even more.
So naturally, investors ask:
Did I already miss it?
Our view is no.
In fact, we believe we are probably still in the second or third inning of the AI data center buildout.
The reason is simple. We are no longer debating whether AI spending is real. We already have proof.
Major hyperscalers including Microsoft, Amazon, Google, and Meta continue increasing spending on AI infrastructure at a rapid pace. Recent industry research showed global data center capital expenditures increased 57% in 2025, with strong spending expected to continue as AI deployments accelerate.
At the same time, electricity demand from AI-related data centers continues to rise sharply, with some forecasts expecting global data center electricity consumption to nearly double by 2030.
This does not look like an industry winding down.
It looks like an industry still being built.
Where We Are in the AI Cycle
| Phase | What Happened | Main Winners |
|---|---|---|
| Phase 1 | AI excitement begins | NVIDIA, Microsoft |
| Phase 2 | Chips and compute boom | Semiconductors |
| Phase 3 (Now) | Massive physical buildout | Power, cooling, fiber optic cable, construction |
| Phase 4 | Enterprise adoption | Productivity software |
This matters because investors often assume they missed the opportunity after seeing the first wave of gains. But many times, the second and third phases are where broader participation happens.
In our view, AI data center stocks may still be entering that stage.
Why We Like the Buildout Better Than Picking One Winner
One of the biggest reasons we like this part of the AI trade is because it removes some of the guesswork.
A lot of investors are trying to answer difficult questions right now. Will OpenAI ultimately dominate? Can Claude take meaningful share? Will Gemini become stronger over time? Those are important questions, but they are also difficult ones to answer with certainty.
The good news is that we do not necessarily have to answer them.
Because regardless of which AI platform ends up leading, the physical infrastructure still needs to be built. Data centers still need electricity, cooling systems, fiber optic cable infrastructure, and ongoing construction. The demand for those things exists whether one chatbot wins or five coexist.
That is what makes this part of the market attractive to us.
Rather than trying to perfectly predict the winning AI platform, we are investing in the ecosystem supporting all of them. In many ways, it feels like a more measured way to participate in AI growth while still maintaining meaningful upside if the buildout continues at its current pace.
What About an AI Bubble?
This is probably the biggest concern investors have, and it is a fair question to ask.
Could parts of AI be overhyped? Absolutely. There is always a chance that expectations get ahead of reality, especially when a new technology grows this quickly. Some companies may disappoint, certain products may not live up to expectations, and valuations in parts of the market could eventually come back down.
But that does not mean the AI data center buildout itself is not real.
This is an important distinction. Even if one AI platform struggles or a specific feature fails to deliver the value investors expected, the infrastructure supporting AI is already being built at an enormous scale. Companies are still committing capital, signing contracts, and expanding capacity because the computing power required for AI remains massive. Industry forecasts continue to point toward sharply higher electricity demand from data centers over the coming years.
In other words, even in a more bearish scenario for AI, it is difficult to imagine these companies simply returning to where they traded several years ago before the buildout began. Growth may slow at some point, and returns may normalize, but the long-term demand for power, cooling, fiber optic cable, and construction still appears very real.
That is why we view these names differently than a speculative software trade. We are not just betting on one AI feature succeeding. We are investing in the infrastructure required to support the entire shift.
Why We Like These Stocks Better Than Guessing the AI Winner
We still like many of the major AI names.
Companies like NVIDIA, Google, and Microsoft have been strong investments and will likely continue playing important roles in artificial intelligence. We continue to have exposure to some of those names.
But when investing in AI, one of the hardest questions is figuring out which platform ultimately wins.
Will ChatGPT continue leading?
Will Gemini catch up?
Will Claude become stronger in enterprise and coding?
Will another platform emerge entirely?
Those are difficult questions, and the truth is, investors do not necessarily need to answer them to benefit from the trend.
Because regardless of which AI platform comes out ahead, the infrastructure still needs to be built.
Every major AI system requires:
- electricity
- cooling and HVAC systems
- fiber optic infrastructure
- construction and maintenance
- physical data center expansion
That is why we like these names.
Rather than betting entirely on one winner, we are investing in the businesses helping power the entire ecosystem. In our view, that creates a more durable opportunity because demand for the infrastructure exists whether one company dominates or several coexist.
In other words, we are not just investing in AI. We are investing in what AI physically requires to function.
The Stocks We’re Watching
Here are some of the names we continue to watch closely:
Corning (GLW)
A key supplier of fiber optic cable and connectivity infrastructure. If data centers keep expanding, fiber demand should continue growing.
Caterpillar (CAT)
A beneficiary of the physical construction side of the buildout, while also remaining a strong industrial business outside of AI.
GE Vernova and Eaton (GEV | ETN)
Power and electrical infrastructure are becoming increasingly important as data center electricity demand rises.
EMCOR Group and Quanta Services (EME | PWR)
These names benefit from installation, engineering, and electrical infrastructure needed to support large-scale development.
We still like companies such as NVIDIA and Google for their role in AI. But increasingly, we also see value in owning the companies helping make the entire ecosystem function.
The Bigger Picture
The AI trade is evolving.
At first, the focus was on the chips. Then the software. Now the conversation is shifting toward what is required to support all of it.
That is why we believe the AI data center buildout deserves serious attention.
Could there be volatility? Absolutely.
But when we look at the amount of spending already committed, the backlog of projects, and the continued growth in power demand, it is hard to argue that this trend is close to finished.
In our view, it still looks early.
Position for the Next Phase of AI Growth
AI is changing quickly, and so are the opportunities around it. At Michael Leslie Investments, we help investors identify where trends are developing and how to position thoughtfully for long-term growth.
Contact Michael Leslie Investments today to discuss how AI infrastructure and data center stocks may fit into your portfolio.
