Latent Space Notes

Where AI research meets the balance sheet — for investors and builders alike.

by Ronald Heller

Joined by 6 readers

Averaging 55 efficiency across 1 mixer

The Second-Order Effects of Cheaper Tokens

The Case for and Against the Capex Supercycle

Jun 26, 2026

Reading the Tea Leaves in Hyperscaler Capex

Reading an AI earnings call is an exercise in separating booked revenue from backlog from ambition. Signed contracts and committed capacity are real; framework agreements and lett…

Jun 22, 2026

Reading the Tea Leaves in Hyperscaler Capex

When a lab ships a new frontier model, the interesting question is rarely whether the benchmark went up. It is whether the price-performance curve shifted enough to unlock a categ…

Jun 18, 2026

Enterprise Adoption Is Slower and Stickier Than the Demos Suggest

The headline number everyone fixates on is training compute, but the margin story is increasingly about inference. As model providers push cheaper, faster variants, the cost of se…

Jun 16, 2026

Enterprise Adoption Is Slower and Stickier Than the Demos Suggest

How to Read an AI Earnings Call

Jun 14, 2026

Datacenter Power Is the Real Scaling Wall

What the Latest Model Release Actually Changes

Jun 7, 2026

Why Memory Bandwidth Is the New Constraint

The Quiet Consolidation of the Chip Supply Chain

Jun 2, 2026