Microsoft and Chevron enter exclusivity deal on powering West Texas AI data center complex

Chevron is developing a 2.5 GW gas-fired power hub in West Texas (scalable to 5 GW) and is in exclusivity talks with Microsoft to colocate an AI campus; the multibillion-dollar project could start coming online as early as late 2027. Chevron has seven GE Vernova turbines ordered and a financial partner, Engine No. 1, and is leveraging Permian output ( >1 million boe/d) to offer behind-the-meter power to hyperscalers. No commercial terms or definitive agreements have been finalized, but the deal signals a strategic pivot by Big Oil into data-center power provision and could ease near-term power constraints for AI campuses.

Analysis

Colocated, behind-the-meter gas capacity creates a new, localized demand center that can reprice regional gas basis and force upstream/midstream capital reallocation. Expect nearby producers to see realized-price uplifts versus distant hubs; that uplift is faster and stickier than headline commodity moves because it bypasses congested pipelines and creates contracted offtake economics for capex-heavy projects.

The supply-chain winners are not just OEMs selling turbines but the recurring aftermarket and services stream that follows large thermal fleets — spare parts, long-duration service contracts, and financing of construction convert one-time equipment sales into 10–20% EBITDA annuities. Conversely, corporate renewable procurement markets will face a near-term pullback in marginal demand where reliability trumps lowest LCOE, pressuring short-duration REC prices and altering hyperscalers’ carbon accounting choices.

Catalysts are binary and path-dependent: contracting/permitting milestones will re-rate equity optionality quickly, while ESG/regulatory shocks (methane rules, local permitting injunctions or carbon policy shifts) can remove that optionality faster than revenue accrues. Net: value crystallizes over years, not quarters — tradeable moves will cluster around deal announcements, turbine deliveries, and regulatory decisions rather than steady-state operations.

Key Decisions for Investors

• Long CVX (2–4% position). Structure as buy-and-call-spread (buy shares + buy 2028 LEAP 1.5–2x out-of-the-money call spread) to capture multi-year optionality without overpaying. Rationale: optionality to monetize regional captive gas demand; exit/trim on definitive long-term offtake contracts or if permitting materially delayed. Risk: regulatory/ESG reversal; stop loss at -15% from entry.
• Long GEV (1–2% position). Purchase 9–18 month call spread to play turbine order backlog and attractive aftermarket margin conversion; target 3x+ payoff if backlogs translate into sustained service revenues. Catalysts: quarterly backlog updates and margin guidance; risk if supply-chain bottlenecks push deliveries beyond near-term windows.
• Long MSFT via 12–24 month LEAP calls (size 1–2%). Fund with sale of short-dated OTM puts to reduce cost. Rationale: energy-secured AI capacity reduces growth bottlenecks and preserves compute expansion optionality; hedge by reducing notional if tech demand cools. Close on meaningful shift in hyperscaler energy strategy away from colocated solutions.
• Pair trade to isolate project optionality: Long CVX / Short XOM (1:0.6 notional). Idea is to express premium for integrated operator execution of power projects while hedging broad oil/gas cyclicality. Trim if macro oil/gas divergence >20% or if project becomes a corporate-wide capital-allocation drag.