2 Data Center Stocks That Could Help Make You a Fortune

Morgan Stanley projects capital expenditure by the five largest U.S. technology companies will reach $1.4 trillion by 2028, supporting sustained demand for AI data-center infrastructure. Iren guides for annualized revenue to rise from $1 billion currently to $4 billion by end-2026 and has begun delivery under its five-year, $9.7 billion Microsoft contract; analysts project $10.7 billion in adjusted operating profit by fiscal 2029. Vertiv reported 24% year-over-year quarterly revenue growth and more than 22% adjusted operating margin, while raising its long-term revenue growth target to 20%-22% annually through 2030 and targeting at least a 27% operating margin.

Analysis

IREN’s valuation hinges less on announced capacity than on whether it can finance multi-gigawatt construction without materially diluting shareholders or levering its balance sheet into a higher-rate environment. The implied fiscal-2029 operating-profit outcome requires sustained premium pricing, high utilization, and on-time energization; any slippage converts a seemingly low terminal EV/EBIT multiple into a capital-intensive development story. Microsoft concentration also creates asymmetric downside: a delayed acceptance milestone or reduced AI infrastructure budget would impair both revenue recognition and the company’s ability to raise project financing on favorable terms.

VRT is a cleaner way to express AI physical-infrastructure demand because cooling and power-density upgrades are required regardless of which cloud operator or GPU platform wins. However, a 37x forward P/E already prices a large portion of the margin-expansion thesis; the next 1-3 months depend on backlog conversion, supply-chain execution, and evidence that liquid-cooling attach rates exceed expectations. The less appreciated beneficiaries of rising rack density may be ETN, NVT and MOD, which offer exposure to electrical distribution, thermal management and enclosure complexity at less demanding expectations than VRT.

The consensus risk is that hyperscaler capex increasingly shifts from shell construction toward internally designed systems and negotiated-volume procurement, limiting pricing power for equipment vendors after the initial AI buildout. Over 6-18 months, power availability—not demand—becomes the gating constraint: developers with contracted, deliverable power retain scarcity value, while unenergized land banks deserve little credit. This thesis is falsified if hyperscaler capex guidance weakens, VRT’s book-to-bill falls below 1x for two quarters, or IREN discloses funding needs materially above operating cash flow and committed project debt.

Key Decisions for Investors

• Maintain a tactical long VRT only on post-earnings confirmation of backlog growth and stable-to-higher gross margin; target a 15-20% upside over 6-12 months, but reduce if forward revenue guidance falls below the long-term growth algorithm or book-to-bill drops below 1x.
• Prefer a diversified AI-power pair: long ETN and NVT versus short a proportional basket of high-multiple data-center developers, including IREN where borrow is available. Hold 3-6 months; the trade benefits if power-distribution content proves durable while development multiples compress on financing or energization delays.
• Do not initiate a core IREN long until project-level capex, debt terms, customer prepayments, and contracted power-delivery dates are independently disclosed. Treat a financing announcement, revised deployment schedule, or customer concentration disclosure as a catalyst/watch item rather than evidence of intrinsic value.
• For existing IREN exposure, use defined-risk downside protection around the next funding or deployment update; a failure to convert contracted capacity into recognized revenue on schedule could produce a 25%+ equity drawdown given the stock’s reliance on terminal profitability assumptions.
• Monitor MSFT and other hyperscaler capex commentary during the next earnings cycle as the leading indicator. A broad capex reaffirmation supports VRT/ETN/NVT over the following 1-3 quarters; a shift toward efficiency-led spend would favor short VRT versus long diversified electrical-equipment peers.