Anthropic and OpenAI Are Both Racing to IPO. GE Vernova Wins Either Way.

GE Vernova reported Q2 2026 orders of $24.2 billion, up 88% year over year, while data-center electrification orders exceeded $5 billion in H1 2026—more than double its full-year 2025 total. The company is positioned to benefit from rising electricity investment by AI leaders Anthropic and OpenAI ahead of their planned IPOs, through its gas-power, wind and grid-equipment businesses. GE Vernova shares remain up roughly 45% year to date and more than 6x since its April 2024 spinoff, although the stock trades at an elevated roughly 38x forward earnings.

Analysis

GEV’s relevant exposure is not merely incremental data-center capex; it sits at the bottleneck between announced compute capacity and deliverable, dispatchable power. The highest-value portion is likely gas generation plus grid interconnection equipment, where long lead times can convert order growth into pricing power and service annuities. This favors GEV over pure data-center hardware vendors if utility queues, permitting, and transmission constraints—not GPU availability—become the binding constraint through 2027-28.

The key equity risk is that investors are capitalizing a peak order-intake cycle as though all backlog converts at current margins. A high-multiple industrial can de-rate sharply if turbine slot availability normalizes, data-center customers defer projects because power purchase agreements are unavailable, or Power segment margin fails to scale with volume due to supply-chain inflation. The important near-term datapoints are backlog conversion, price/cost performance, working-capital consumption, and the mix of firm orders versus framework agreements—not headline data-center commitments.

Second-order beneficiaries include ETN, HUBB and PWR: electrification gear, transmission equipment, and engineering/construction may capture more spend per incremental MW than generation OEMs once generation contracts are placed. Conversely, merchant power names can face a two-sided outcome: tighter regional capacity lifts realized prices initially, but rapid new gas additions ultimately cap scarcity rents. The consensus appears too focused on OpenAI/Anthropic financing events; utility procurement, interconnection approvals, and turbine delivery schedules determine actual revenue timing.

The setup is constructive structurally but offers limited margin for execution misses over the next 1-3 months. Prefer buying dislocations around earnings or order volatility rather than chasing AI-related headlines; a sustained slowdown in electrification orders or downward revision to 2027 free-cash-flow conversion would falsify the premium-multiple thesis.

Key Decisions for Investors

• Maintain a measured long GEV only on a 10-15% post-results pullback or after verified backlog-to-revenue conversion improves; target a 12-18 month hold, with risk defined by a material miss to Power/Electrification margin or FCF guidance. Avoid adding solely on AI IPO headlines.
• Pair trade for 6-12 months: long ETN / short GEV in equal dollar amounts if GEV’s valuation premium widens further. ETN has cleaner exposure to electrical distribution and data-center power architecture, while the short leg hedges turbine-cycle and execution risk; exit if GEV demonstrates sustained margin expansion beyond guidance.
• Add PWR to an AI-power infrastructure basket on confirmed utility transmission awards, not announced data-center projects. Its catalyst path is 2-4 quarters later than equipment orders but has less dependence on gas-turbine supply; risk is permitting delays and labor-cost inflation.
• Monitor disclosed GEV turbine delivery slots, electrification book-to-bill, backlog cancellation language, and utility interconnection queues each quarter. A deterioration in any two metrics is a signal to reduce exposure before consensus earnings estimates reset.