AI Capex Flow MapHyperscaler → Supplier Attribution
All suppliers
Power & ElectricalSEC filings

GE Vernova (GEV)

Gas turbines (HA-class), grid equipment, electrification for DC power

Modeled AI capex flow
$13.9B

from Big 5 forward AI-attributable spend

Hyperscaler exposure
20.0%

of revenue (latest 10-K)

Top hyperscaler customer
AMZN

modeled $4.0B flow

Total backlog
$116.0B

Q4 2025

Forward visibility: Total backlog

The leading indicator AM analysts read as capex commitments convert to supplier revenue.

Gas Power + Grid + Electrification + Wind. Heavy-duty gas turbine slots booked into 2028+. Behind-the-meter DC interconnect demand a key driver.

As of: Q4 2025Source filing
Flow attribution by hyperscaler

Modeled dollars from each hyperscaler's AI-attributable forward capex.

CustomerModeled flow% of total
Amazon (AMZN)$4.0B28.5%
Microsoft (MSFT)$3.2B23.3%
Meta Platforms (META)$2.8B20.3%
Alphabet (GOOGL)$2.7B19.4%
Oracle (ORCL)$1.2B8.5%
Computed from disclosed customer concentration and per-hyperscaler accelerator preferences. Per-edge values can be reconstructed from the Methodology page.
Notes & source

Multi-year turbine backlog, lead times pushing to 2028+. Indirect AI capex play via behind-the-meter gas.

View SEC filings page