Barclays’ Top Gas & Midstream Stock Picks for the AI Infrastructure Boom

Barclays projects AI infrastructure spending from hyperscalers and AI labs could exceed $1 trillion annually before peaking in 2028, implying more than $300 billion above current consensus and boosting demand for natural gas infrastructure. The article highlights Exxon, Chevron, Enbridge, Williams, Enterprise Products Partners, Kinder Morgan, Energy Transfer, TC Energy, Cheniere, and ONEOK as key beneficiaries, with several also reporting beats, guidance raises, dividend increases, or analyst upgrades. The setup is constructive for gas producers and midstream operators, though stock-specific reactions will depend on individual earnings, guidance, and rating changes.

Analysis

The market is starting to price AI infrastructure as a real commodity demand shock, not just a software capex story. The underappreciated second-order effect is that power, gas transportation, and LNG molecules become the gating factor, which shifts bargaining power toward regulated or quasi-regulated midstream names with long-duration contracts and away from more sentiment-sensitive “AI beneficiaries” like Microsoft, where the capex overhang and OpenAI dependency create a valuation ceiling even after a clean quarter.

Among the energy complex, the best risk-adjusted winners are not the highest beta producers but the toll collectors with embedded volume growth and limited commodity exposure. Midstream operators with expanding pipe capacity and export adjacency should see a multi-quarter re-rating as AI load growth translates into incremental baseload demand; that argues for ENB, KMI, TRP, ET, and OKE over pure upstream names. LNG-linked assets add an extra layer of optionality because they can monetize both domestic power demand and global supply tightness, but they are more exposed to policy, permitting, and execution risk.

The main risk is timing: AI data-center demand is a multi-year thesis, while consensus can still fade if hyperscaler capex slows for even one or two quarters. If the market starts to worry that AI training becomes less power-intensive due to model efficiency gains or slower recursive scaling, the most crowded “infrastructure winner” trades can compress quickly. Conversely, if sovereign AI and China accelerate the buildout, this becomes a capacity cycle with a much longer runway than most energy investors are modeling.

The contrarian miss is that this is not just bullish gas prices; it is bullish midstream utilization and contract extension economics. That makes the best longs the companies with visible takeaway expansion and dividend support, while the weakest link is any name whose upside already embeds a straight-line demand boom. Microsoft may remain structurally strong, but near-term the asymmetry is better expressed in the energy infrastructure chain than in the AI platform leader itself.

Key Decisions for Investors

• Long KMI or TRP vs short MSFT on a 3-6 month horizon: express the view that AI capex capex intensity supports infrastructure beneficiaries while MSFT faces multiple compression from rising capital intensity; target a 10-15% relative spread if hyperscaler spend stays elevated.
• Initiate a basket long ENB/EPD/ET/OKE, sized equally, for 6-12 months: these names should capture volume growth with lower commodity beta; use a 7-10% drawdown stop, expect high-single-digit total return plus dividend carry.
• Buy LNG on pullbacks for a 6-9 month trade: higher-conviction beneficiary if the market re-rates export optionality from power demand and global LNG scarcity; risk/reward improves on any broad energy selloff, with upside tied to renewed export margin strength.
• Prefer EPD and KMI over CVX in this theme: CVX has only partial exposure to the AI power thesis, while midstream tolling should re-rate faster; pair long EPD / short CVX for a cleaner expression of gas throughput over upstream commodity risk.
• Avoid chasing MSFT strength into earnings reacceleration until capex commentary stabilizes: any rally driven by AI enthusiasm is vulnerable if investors focus on free-cash-flow dilution; use rallies to trim or hedge with short-dated puts if capex guidance stays aggressive.