Is the AI hype cooling down?
Live signal index
Is the AI hype cooling down?
❄ Cooling
Overheating 🔥
A composite reading of physical build-out and platform economics. Higher means the<br>expansion is still intensifying; lower means the momentum is fading.<br>Here's what goes into it.
See the breakdown
What the answer is built from
Five independent signals, each weighted by how directly it reflects real, committed<br>demand. Each is scored from 0 (contracting) to 100 (surging).
The trajectory
Where the money and the megawatts are actually going.
Combined infrastructure spend, per quarter
Billions of dollars
Spend by operator
$B / quarter
New data-center power queued
Gigawatts / year
Total new power capacity queued
Gigawatts / year
Where data-center load is landing
GW in queue, top regions
Where it could actually jam
Demand is only half the story. Even with the money flowing, the build-out runs into hard<br>physical limits — and four of them govern how fast AI compute can be poured. Here's how<br>binding each becomes year by year through 2030, and the companies with the most leverage<br>over each chokepoint.
How binding<br>easing<br>tight<br>severe<br>binding<br>· hover a year for detail
Company names indicate supply-chain exposure to each bottleneck — illustrative, not investment<br>advice. Trajectory synthesized from IEA, McKinsey, SemiAnalysis, SK Group and industry<br>commentary (2024–2026).