3 Nuclear Energy Stocks Powering the AI Boom in August

The AI power buildout thesis remains supportive: Constellation (CEG) reported Q1 2026 revenue of $11.12B (+63.9% YoY) and raised/confirmed 2026 adjusted EPS guidance to $11.00–$12.00, targeting >20% base EPS growth through 2029 while citing multi-decade Microsoft-style PPAs tied to hyperscaler loads. Cameco (CCJ) benefited from elevated uranium prices (spot $88.49/lb, +34% YoY) with Q1 revenue missing consensus by 25.6% ($606.3M vs $815.1M) but EPS of $0.34 matching estimates, and 2026 guidance remaining intact amid high valuation risk (forward P/E 77x). BWX Technologies (BWXT) delivered a stronger Q2 with revenue $901.6M (+18% YoY) and non-GAAP EPS $1.07 (vs $1.042 expected), raising full-year guidance to ~ $3.80B revenue and $4.70–$4.80 EPS, with backlog at $8.65B—though risks include nuclear/utility policy uncertainty and potential government shutdowns for BWXT’s government segment.

Analysis

The market is increasingly treating AI power as a supply-chain bottleneck rather than a simple utility story. The best equity expression is not the most obvious commodity beta; it is the names that can lock in long-duration contracted cash flows or remove bottlenecks in nuclear manufacturing, because those are the pieces that convert capex into visible earnings. That favors CEG and BWXT more than a pure uranium lever.

CCJ still works if uranium remains tight, but it is the most consensus-owned leg of the trade and therefore the most vulnerable to valuation compression if spot simply stalls instead of exploding higher. The key second-order risk is that long-term fuel demand can stay constructive while equities underperform because contract visibility is already embedded and the market starts paying attention to execution, tax, or regulatory friction instead of the commodity narrative.

The contrarian miss is timing: hyperscaler demand does not automatically translate into immediate nuclear cash flow because interconnect, permitting, and transmission are the real bottlenecks. In the next 1-3 months, the trade is more about sentiment and order flow; over 6-18 months, the thesis only works if data-center load additions keep outpacing grid buildout and if nuclear project execution stays clean. The main falsifier is any sign that load forecasts or contracting cadence slow enough to push monetization into later years, which would cap multiple expansion and punish the higher-multiple names first.

Key Decisions for Investors

• Go long CEG on pullbacks into the mid-$250s for a 1-3 month catalyst trade, targeting a move back toward the low-$300s if contract headlines keep coming; reduce risk if management trims 2026 EPS or the stock loses the low-$240s.
• Prefer BWXT over CCJ for the next 3-6 months: long BWXT / short CCJ as a pair to isolate the pick-and-shovel benefit from commodity and balance-sheet risk; target 10-15% relative outperformance if nuclear supply-chain bottlenecks stay tight.
• Do not chase CCJ at current levels; if owning the theme, use smaller sizing or call spreads rather than outright equity because a flat uranium price can still trigger multiple compression over the next 1-2 quarters.
• Add BWXT on any post-earnings consolidation rather than breakout buying; the thesis breaks if government funding noise hits backlog conversion or if the market starts discounting slower commercial nuclear capex in 2027.