EA://INTEL — Why Six and Twelve Points Shape Meaning On A Sphere

Part two argues that when meanings must be maximally distinguishable, they are forced into the shape of specific solids rather than chosen by taste.

An AI system needs a way to pack many distinct ideas into a fixed space without them blurring together. That packing problem is not abstract art; it is a constraint on how meaning can be stored if you require that no two states be confused. If you ask a machine to place six points on a sphere so the nearest neighbors are as far apart as possible, the system searches for the arrangement that scores best on that measure and finds the octahedron. Twelve points converge on the icosahedron under the same rules. An analogy: think of placing magnets on a balloon; if you push them apart until they stop moving, they settle into those shapes automatically.

We ran this search from many random starting positions and it always converged to exactly those solids, matching published results that were already known. The consequence is strong because any mind, anywhere, that wants six or twelve maximally distinguishable states is pushed toward the same shapes by the mathematics of the sphere itself. This forced structure becomes a focal point that strangers can share without negotiating anything beforehand.

Between parts: if two AI systems needed a shared reference and could not talk first, would a forced structure like this serve? Tell me where the idea breaks.