Meta's massive nuclear power deals will help US 'win' AI race against China, executive says

Meta has signed 20-year agreements to purchase nuclear power from Vistra’s Perry and Davis-Besse plants in Ohio and Beaver Valley in Pennsylvania and is supporting development of small modular reactors with Oklo and TerraPower, potentially supplying up to 6.6 GW by 2035. Executives say the capacity will underpin surging data-center demand and bolster the U.S. position in the AI competition with China; the deals (plus an earlier Illinois extension) could make Meta one of the largest corporate buyers of U.S. nuclear power and triggered notable equity reactions (Vistra +10.56%, Oklo +7.90% reported).

Analysis

Market structure: Meta’s 20-year PPAs and 6.6 GW target to 2035 directly benefit large nuclear operators (Vistra/VST) and SMR developers (Oklo/OKLO, TerraPower) by securing long-duration contracted revenue and catalyzing supply-chain investment. Expect upward pricing power for merchant nuclear owners (higher equity valuations, lower credit spreads) but modest absolute demand impact — 6.6 GW ≈ <1% of U.S. nameplate capacity, so winners are concentrated, not systemic.

Risk assessment: Key tail risks are regulatory delays (NRC licensing), CAPEX overruns, and political reversal (state-level permitting or subsidy cuts) that can push project timelines 3–7+ years and double costs. Short-term (days–months) see equity re-rating and volatility; medium-term (1–3 years) depends on financing/permits; long-term (to 2035) depends on SMR commercialization and grid interconnection.

Trade implications: Direct plays: buy Vistra equity and selective uranium/mining exposure while using option structures to manage execution risk; avoid broad utility longs that own gas-heavy fleets vulnerable to dispatch loss. Use pair trades: long VST vs short gas-fired regional utilities or UNG to express nuclear replacing peaker gas; prefer staged allocation with licensing-based scale-ins (NRC milestones, DOE awards).

Contrarian angles: The market narrative overstresses near-term AI-driven electricity demand — 6.6 GW by 2035 is small vs AI power needs and will not materially change macro power markets short-term. History of large infrastructure PPAs shows long lead times; pricing for uranium/miners and speculative SMR developers may be overbought after the initial pop. Hidden negative: transmission bottlenecks and skilled-labor inflation could erode project returns.

Key Decisions for Investors

• Establish a 2–3% portfolio long position in VST (Vistra) within utilities allocation: scale in over 6–12 weeks, target 20% upside in 12–18 months, stop-loss at -15% from entry; increase to 4–5% only after NRC/DOE milestone (license approval or firm construction financing) confirmed.
• Allocate 1–2% to uranium exposure (URA ETF or CCJ): dollar-cost average over 6–24 months; add incremental 0.5% if uranium spot rises >15% or if multiple SMR projects receive construction permits, target 25–35% gain over 12–36 months.
• Implement a relative-value pair: go long VST (2%) and short 1.5% exposure to a gas-heavy regional utility ETF or UNG (synthetic) to hedge power-dispatch risk; rebalance at each quarter or when spread narrows/widens by >200 bps.
• Take a 0.5–1% speculative position in OKLO (OKLO) equity or listed derivative: small allocation due to execution risk; use long-dated (12–24 month) calls rather than spot equity to limit downside, and exit or scale up only after public licensing/first commercial deployment confirmation within 12–36 months.