Skip to main content

Lab Notebook · The Two-Organ Mind

The two-organ mind — deriving an architecture from a dead speed thesis

Out of the navigator's wreckage (incompressible weights, a masking ceiling around 2.6×) came a different idea: a mind doesn't need one organ that both knows and checks — it needs two. A small model as the gate pointed at the mind. A toy proof of the mechanism ran in ~47 ms; for CODE the two organs already exist — the language and its type-oracle supply structure, the compiler-verification gate supplies exact meaning. Dated across BS053-056, July 2026.

JR
Jon RossFounder, vocabotics — 15 years building safety-critical systemsLab report · dated 10 July 2026
Verified by a human. Drafted with AI, verified by a human. Jon Ross, 10 Jul 2026
Living document. Reviewed 10 Jul 2026
Entry date
10 July 2026
Category
Language
Lead over the world
a derived architecture, not yet a shipped lead
Access
🔒 Protected
ResearchPartly proven

The micro model is the gate pointed at the mind — not a smaller version of it.

~47 ms
toy proof of the two-organ mechanism, measured end to end
language + type-oracle → gate
for CODE the two organs already exist: structure from the language, exact meaning from the gate
BS053 → BS056
the council brainstorms across which the architecture was derived from the navigator's wreckage
hardcoded
the toy's long-range rule — not learned, not yet generalised

Honest evaluation

Partly proven

The toy mechanism is measured and real at ~47 ms; for CODE, both organs (language + type-oracle as structure, the gate as exact meaning) already exist and run. Scaling the architecture to a full mind, and the toy's hardcoded rule becoming a learned one, are both still open.

What we learned — a preview

The honest edges are open even when the engine is not. The full report — what it was, what we built, and the measured internals — is protected.

  • The toy proof's long-range rule was hardcoded by hand, not learned or derived from data. The ~47 ms result shows the two-organ mechanism can run fast; it does not show the mechanism can learn or generalise a rule on its own.
  • Scaling this from a toy to a navigator-sized mind is a stated hypothesis, not a measured result. Nothing here claims the 2.6× masking ceiling or the incompressible-weights finding are solved by this architecture — they are the reason it was tried.
  • For domains other than code, the 'gate' side of the pair does not yet have an equivalent to the type-oracle's exactness. Code is the case where this architecture is furthest along, not the general case.
Protected report

The deep body of “The two-organ mind — deriving an architecture from a dead speed thesis” is behind access.

We open the demos, the specs, and the method; we protect the engines, the model internals, and anything that touches commercial, safety-critical work. This report describes an engine — so its details are gated, and no client, company, or project is named. Members can read it in full; if you have a genuine reason to see it, tell us who you are and why.

Proofs & sparks

We demonstrate rather than assert. Each ✅ proof is a visible result with a hard figure.

  • A toy two-organ mind, measured~47 ms toy proofa toy proof of the two-organ mechanism — a small, cheap gate checking a separate producer — ran end to end in ~47 ms; for code, the two organs (the language + type-oracle as structure, the gate as exact meaning) already run in production.

Where this connects

    We use cookies.