AI infrastructure stocks span far more than the chipmakers that grab headlines. Before a single graphics processor can train a model, someone has to build the building, run the power into it, and keep it cool, and each of those steps is a business with publicly traded exposure. The American Prosperity Report’s AI Powerhouse pitch centers a financing giant in that stack, and understanding the full chain is the fastest way to judge whether the pitch makes sense.
The Full Stack, Top to Bottom
Think of AI infrastructure as a stack with four layers. At the bottom is power: generation, transmission, and the renewable and grid capacity that keeps a facility running. Above that is the physical shell: the land, the building, and the cooling systems that stop thousands of servers from overheating. Above that is the computing equipment itself. And wrapped around all of it is the capital layer, the money that finances every stage of the buildout.
Most AI headlines live at the computing-equipment layer. The American Prosperity Report pitch is deliberately aimed one layer down and one layer out: at the power and the physical infrastructure, and at the asset manager that finances both. That is a meaningful difference, because each layer carries its own economics and its own risk.
The Power Layer Comes First
Data centers are power-hungry in a way few industries can match. A single modern facility can draw as much electricity as a small city, and the constraint on new AI capacity is increasingly not the chips but the power available to run them. That is why the buildout is pulling in generation of every kind, from natural gas and nuclear to wind and solar, and why grid interconnection queues have become a bottleneck.
We dig into the electrical side of the same trend in our data center power demand explainer. The short version is that whoever controls dispatchable power near a data center campus holds a scarce asset, and scarcity is where pricing power lives.
The Cooling and Building Layer
The building itself is less glamorous but no less essential. Servers generate enormous heat, and removing that heat is a hard engineering problem that grows with every new generation of denser chips. Liquid cooling, purpose-built floor layouts, and backup power all add cost and complexity, and they reward operators who can build fast and run efficiently.
This layer is also where the financing structure matters most. Data centers are capital-intensive and slow to pay back, which is why they attract long-dated capital from pension funds, insurers, and sovereign wealth funds rather than venture money. That preference for patient capital is a direct handoff to the asset managers.
The Capital Layer
The capital layer is where Brookfield lives. It is not a chipmaker or a builder; it is a global alternative asset manager with roughly $1.3 trillion in assets under management that raises money from institutions and deploys it into infrastructure, including data centers and the power behind them. The AI boom shows up in its business as a demand wave, not as a product.
We explain how that fee-driven model actually makes money in our alternative asset managers explainer, and the broader set of ways to invest in the buildout is covered in our AI data center investment guide. The key point for an infrastructure-stock investor is that the capital layer earns fees on the whole stack rather than betting on a single project.
What the Pitch Actually Claims
The American Prosperity Report presentation wraps the stack in some eye-catching numbers. It cites a “$7 trillion pipeline” of AI-infrastructure deals, a “40% undervalued” read on Brookfield, and a “1,800% gain over 10 years” framing, and it leans on Bill Ackman’s Pershing Square having expanded its Brookfield stake. The first of those is directionally grounded in real hyperscaler commitments from Google, Microsoft, Meta, NVIDIA, and the US government; the return and valuation claims are marketing arithmetic that deserves independent checking.
The honest read is that the stack is real and growing, but no layer of it is a secret. The question for any investor is which layer matches their risk tolerance, and what they are paying for it today.
Reading the Stack Honestly
Two cautions apply to the capital layer specifically. Brookfield carries roughly $270 billion in consolidated debt, most of it non-recourse to the parent, and part of its portfolio sits in commercial real estate that has been under stress. Those are balance-sheet facts, not fatal flaws, but they mean the stock should be sized like an infrastructure holding rather than a momentum tech trade.
The broader point is that AI infrastructure is a multi-year buildout with real cash flows, not a single-stock lottery ticket. Investors who map the stack before they buy are in a better position than those who buy the story first and ask questions later.
Ready to see the research? Click here to access Charles Mizrahi’s report.
NewsletterVetter is an independent publication. We receive compensation from some of the services we review through affiliate links. Nothing on this site is investment advice. Always do your own research.