Skip to content
Technology How it works Breeder — Hyperion Burner — Aegis Burner — MetroVolt AI-Native Architecture Magnets Fuel cycle Safety Roadmap
Solutions AI & Data Centers Defense & Government Grid & Baseload Neutron Detection Quantum
Learn Technical Library
Proof Publications Whitepapers Technical Library Open Science & Reproducibility The Honest Gates
Company About / Mission Leadership Environment Health & Safety Investors Careers Press Contact
3D Model
Technical LibraryValidation & Reproducibility
Technical

What Reproducibility Does and Doesn't Prove

Reproducing a computation proves it follows from the model and inputs — not that the unbuilt machine will behave that way.

Validation & ReproducibilityUpdated 2026-08-11

An honest account of the limits of reproducibility: re-running a Kronos deposit proves the published number follows rigorously from the stated model and inputs. It does not, by itself, prove that a machine — not yet operating, though the HYPERION breeder is scheduled to begin construction in Q2 2027 — will match the model. Kronos states both halves of that clearly.

This is why every figure is labelled a computed design target, and why the binding gates are named separately.

Two different claims'The math is right' and 'the machine will work' are different claims; Kronos never lets the first stand in for the second.

Common questions

What does reproducibility actually prove?

That a published number follows correctly from the stated model and inputs — that Kronos did the calculation it says it did. Anyone can re-run the open engines and get the same result.

What does it not prove?

That an unbuilt machine will behave exactly as modelled. That is why every figure is a computed design target, and why physical gates and high-fidelity analysis remain on the path to build.

← PreviousTier-1 and Tier-2 Reproduction
Content reviewed August 2026 · design-and-simulation stage