Core Scientific plans 1.5 GW expansion at Texas data center By Investing.com

Core Scientific plans to expand its Pecos, Texas campus to about 1.5 GW of gross power capacity, or roughly 1.0 GW of leasable power, with initial capacity expected in early 2027. The company also secured an additional 300 MW under contract and bought more than 200 acres to support the AI data center buildout, while continuing to shift away from bitcoin mining. Separately, it priced $3.3 billion of senior secured notes due 2031 and added a $500 million JPMorgan commitment, though it remains unprofitable despite analysts projecting 96% revenue growth this year.

Analysis

CORZ is no longer behaving like a pure bitcoin proxy; the market is beginning to re-rate it as a power-constrained AI infrastructure developer with an embedded real estate and grid-option value. The second-order implication is that the balance sheet becomes the real product: once a hyperscaler-quality site is on the power queue, cheap capital matters more than operating margins, which is why the new debt package is strategically important even if it adds leverage. This also puts pressure on smaller colo developers without captive power rights, because the scarce resource is not GPUs but interconnect-ready megawatts.

The key timing issue is that this is a 2027 story, not a 2026 earnings story. Near-term revenue still depends on the declining crypto base, while the market is likely to look through the current unprofitability only if management can show firm pre-lease commitments and staged capex discipline. If execution slips by even two quarters, the equity could de-rate quickly because the stock is already pricing in a fairly clean conversion from mining asset to AI campus.

The contrarian view is that investors may be overpaying for power optionality before proving tenant demand at economic rents. A 1 GW leasable pipeline sounds enormous, but AI infrastructure customers are increasingly sensitive to delivered cost per kW and deployment certainty; if industry supply accelerates in 2026-27, lease-up could normalize faster than bulls expect. The hidden risk is capital intensity: every incremental MW developed raises refinancing and completion risk, so the equity trade may eventually become more about credit spread than about narrative momentum.

On the other hand, the financing itself can become a catalyst for a squeeze if the market interprets it as de-risking the buildout and validating asset value. The stock can keep working for months if management continues converting non-core mining assets into contracted HDC revenue, but the path is likely volatile around any update on pre-leasing, draw schedules, or power delivery milestones. The setup favors investors who can tolerate headline-driven swings but need a hard stop if the company starts funding expansion faster than it can sign customers.

Key Decisions for Investors

• Long CORZ on pullbacks over the next 2-6 weeks, targeting a continuation move as the market prices in AI optionality; trim aggressively if management fails to show pre-lease progress or if financing terms widen materially.
• Use a call spread in CORZ 6-12 months out rather than stock for asymmetric exposure to power-asset re-rating; structure around the next site/construction update so theta decay is not the main risk.
• Pair trade: long CORZ / short a legacy miner basket over 3-9 months, betting that scarce power and AI conversion assets outperform pure bitcoin beta as capital rotates from hash-rate to infrastructure.
• If available, buy CORZ credit protection only as a hedge against execution slippage; the equity thesis is long-dated, but the first failure mode is leverage + delay, not demand collapse.
• Watch for a tactical long in listed power-infrastructure peers if CORZ validates leasing economics; the trade works best if the market starts paying up for megawatt control rather than current earnings.