Today · post October 14

Heim says 20x compute buys only months; it's his estimate

Compute-governance researcher Lennart Heim says a 20x compute advantage buys only about six months of lead in model capability. If distillation is blocked as well, the lead is about 12 months. He wrote: "Looks like a 20x compute advantage only gives you ~6 months on model capabilities. Efficiency exponentials are real." (Lennart Heim, 2026-10-14) The 12 months comes from the next post in the same thread. Distillation is training a smaller or rival model on a stronger model's outputs. His reasoning is one sentence: the exponential gains in efficiency are real. On that account, the side that is behind can catch up with far less compute. Both numbers are his estimates, and the posts attach no underlying analysis. Our reading: if his estimate holds, the lead that export controls buy is measured in months, and blocking distillation can double it. How we checked →

Concept illustration on a pale orange background: a black road with white dashed lines runs across the middle, carrying two wheeled carts, one behind the other, each holding an open laptop whose screen shows a speech-bubble icon, with a short red stretch of road between them. The front laptop on the right is joined by a thick cable to a row of four large server racks at upper right, and the rear cart on the left is joined by a thin wire to a single small server cabinet below the road; large black quotation marks sit at left, and a hand in a dark sleeve holds a smartphone with a white screen in the lower-left corner.

Source and checks

This item ran in the brief of Oct 15, 2026, alongside 18 other items. How we checked it, the quote checks and what was still missing that day are on that issue's How we checked page.

Cite this item

SecondSource (2026). "Heim says 20x compute buys only months; it's his estimate." https://secondsource.io/en/expert/2026-10-15-1/. Accessed access date.