For 20 years, commercial real estate (CRE) was supposed to be the boring diversifier. Tech melted down? Office leases and apartments chugged along.

That relationship is breaking.

Data centers are turning CRE into a high-beta derivative on the AI arms race—and the numbers are getting big enough that the “office building economy” is quietly being replaced by the “server-farm economy.”

So yes—real-estate investors are getting rich.

But here’s the part most people still don’t understand:

The best “second-order” winners aren’t the landlords. They’re the companies that get paid to build the digital factories.

If you’ve never toured a modern AI data center, picture this:

An office tower is a box for humans.
A data center is a power plant + refrigeration system + high-security vault that happens to contain servers.

That matters because the value doesn’t just go into land and walls—it explodes into:

McKinsey explicitly highlights how much of the spend is tied to mechanical/electrical systems—procurement and installation of MEP can exceed $250B by 2030 as demand scales.

And JLL’s 2025 read on the market shows why suppliers love this: big pipelines, heavy preleasing, and tight vacancy that keeps the build machine running.

Traditional CRE is diversified by tenant type: a building has law firms, insurers, consultants, etc.

Data centers are not that.

They’re increasingly dependent on a narrow set of hyperscalers and AI-driven demand. And the deal structures are getting more financial-engineered—big enough that they’re pulling in private credit and SPVs.

Example: Meta’s Louisiana Hyperion site financing (reported by Reuters) involved $27B+ of debt and ~$2.5B of equity in an SPV, with Meta as developer/operator/tenant—but not the borrower. That’s a neon sign that “AI real estate” is now its own capital-markets ecosystem.

Translation:
It’s no longer “buy a building, collect rent.” It’s “finance an industrial megaproject that must hit performance specs.”

That’s why the cleanest way to ride the boom can be upstream—the people selling the required hardware, services, and labor regardless of which landlord wins the lease.

Not advice—this is the “who gets paid when shovels hit dirt” map.

If power is the new real estate, these are the companies building the on-ramps.

Why they win: even if AI demand “slows,” the grid backlog doesn’t magically disappear. Data-center power requirements force upgrades that take years.

This is the high-margin plumbing inside the campus—where data centers differ most from offices.

Why they win: data centers don’t “value engineer” power reliability. Downtime is existential.

AI racks are hotter, denser, and increasingly push liquid cooling and more complex thermal systems.

Why they win: AI doesn’t just require chips—it requires heat removal. Cooling becomes a gating item, not a nice-to-have.

Even the best interconnection doesn’t eliminate redundancy requirements.

Why they win: data centers are engineered for uptime—backup power is not optional.

These are the firms that actually show up and install the systems—often with backlog visibility and pricing power during booms.

Why they win: in a buildout, equipment is important—but skilled labor and execution capacity become the choke point.

If you’re pouring pads and building massive shells, the “dumb” inputs move a lot of dollars.

Why they win: you don’t get “AI scale” without physical scale.

This is adjacent to construction, but it matters because it determines whether projects actually get built.

Why they win: if banks and public markets get tighter, private capital often fills the gap—at a price.

Unlike office projects, data centers have technical performance and timing constraints. Miss the power, miss the schedule, miss the redundancy spec—and economics change fast.

The market can finance buildings faster than utilities can build generation/transmission. That mismatch creates “starts and stops,” and that’s when developers get hurt.

Data centers feel like real estate, but they trade like AI infrastructure. If AI capex sentiment turns, data-center risk premiums can gap wider (even if leases are long).

If you want to play the “data centers are replacing offices” super-cycle without having to underwrite which landlord wins which hyperscaler:

Own the toll roads.
Power gear, cooling, contractors, grid buildout.

Those businesses get paid if the AI arms race is euphoric…
…and they still get paid if the arms race just becomes “baseline infrastructure” that governments and utilities have to build anyway.

Here’s the headline Wall Street is acting on right now:

Fear is gone.
The VIX is back below the “everyone relax” line, and the tape is behaving like we’re sliding into 2026 with a soft landing, a Santa rally, and a clean runway.

But here’s the part most people miss:

Low volatility doesn’t mean low risk. It often means risk is being priced in the wrong place.

1) Complacency is back… and it’s cheapening the price of insurance.
When the VIX is pinned low, the market is basically saying: “Nothing bad is likely to happen soon.”
That’s exactly when hedges are cheapest, and exactly when people stop buying them.

The practical implication isn’t “sell everything.” It’s this:

2) The AI trade is no longer a single trade — it’s splintering.
The 2023–2025 regime was easy: “AI = up.”
The 2026 regime is messy: AI winners and losers separate based on balance sheets, funding, utilization, and monetization.

That creates a weird setup where:

That’s how you get low VIX + high dispersion. That’s not “safe.” That’s “selective.”

3) The economy is quietly more dependent on AI capex than people want to admit.
When AI-related investment categories represent an outsized share of real growth, you’ve effectively built a mini “capex engine” inside GDP.

Implication:

That’s not theoretical. It’s exactly how “one sector” turns into “macro” when it becomes big enough.

4) Sentiment indicators are starting to lean euphoric.
BofA’s positioning/sentiment measures have crept high enough to trigger the classic contrarian read: crowded optimism can be fragile optimism.

This doesn’t mean a crash is imminent.
It means the market is less shock-absorbed than it looks.

1) “VIX low = guaranteed melt-up.”
Low vol can persist… until it doesn’t. The VIX is a current mood indicator, not a future event predictor.

2) “AI is over.”
If AI is still a meaningful contributor to growth and capex, then the theme isn’t dead — it’s maturing into a financing/ROI story. The bubble talk is loud. The build-out is still real.

3) “Broadening breadth = no more risk.”
Broader participation is healthy… but it can also mask the fact that the most crowded trades are still brittle.

But 2026 is shaping up to be a year of fat tails:

Low VIX doesn’t cancel those. It just means the market isn’t paying for them right now.

Not advice — just the framing:

Today’s stocks is Monolithic Power Systems (MPWR)…..

Monolithic Power Systems (MPWR) is one of the cleanest “AI infrastructure picks-and-shovels” names as the bottleneck shifts from compute availability to power delivery and efficiency. As racks and clusters get denser, every incremental watt pushed into GPUs/accelerators creates a cascading requirement for tighter regulation, higher conversion efficiency, better thermal behavior, and more complex multi-rail architectures—exactly where MPWR’s high-density power management and integrated module approach tends to win share and expand content-per-rack. The key asymmetry is that this is not a one-cycle GPU story: as hyperscalers move toward higher-power rack-scale designs (and eventually more standardized, higher-voltage architectures), the power “bill of materials” rises structurally, and MPWR’s attach opportunity compounds with each generation. If 2026 is the year the market stops asking “who has the fastest chip?” and starts asking “who can power this reliably at scale?” MPWR belongs near the top of that list.

My recent discussion about the implications of AI…..

The H.E.A.T. (Hedge, Edge, Asymmetry and Theme) Formula is designed to empower investors to spot opportunities, think independently, make smarter (often contrarian) moves, and build real wealth.

The views and opinions expressed herein are those of the Chief Executive Officer and Portfolio Manager for Tuttle Capital Management (TCM) and are subject to change without notice. The data and information provided is derived from sources deemed to be reliable but we cannot guarantee its accuracy. Investing in securities is subject to risk including the possible loss of principal. Trade notifications are for informational purposes only. TCM offers fully transparent ETFs and provides trade information for all actively managed ETFs. TCM's statements are not an endorsement of any company or a recommendation to buy, sell or hold any security. Trade notification files are not provided until full trade execution at the end of a trading day. The time stamp of the email is the time of file upload and not necessarily the exact time of the trades. TCM is not a commodity trading advisor and content provided regarding commodity interests is for informational purposes only and should not be construed as a recommendation. Investment recommendations for any securities or product may be made only after a comprehensive suitability review of the investor’s financial situation.© 2025 Tuttle Capital Management, LLC (TCM). TCM is a SEC-Registered Investment Adviser. All rights reserved.