Intelligence and Stupidity as Universal Properties of Rule Systems

This framework redefines intelligence and stupidity as scale-free, substrate-independent properties of any computational system. Intelligence is the deployment of a rule system that efficiently maps inputs to correct outputs — minimizing computational cost relative to the problem's inherent complexity. Stupidity is not the absence of knowledge but the application of a rule system that is provably self-defeating: one guaranteed in principle to produce wrong answers regardless of how long it runs. Ignorance — mere data insufficiency — is categorically distinct because it is recoverable.

The Rubik's Cube serves as a canonical illustration. The "God's number" result proves that every configuration is solvable in 20 moves or fewer, establishing a theoretical ceiling on optimal performance. A solver restricted to rotating a single face, however, occupies a region of the solution space from which the goal state is unreachable — not practically difficult but formally impossible under that rule set. The distinction between these two approaches is not one of degree but of kind.

The universality claim is the most provocative element. Bacterial chemotaxis, swarm coordination in fish shoals, market-clearing mechanisms in economies — each can be evaluated by whether its operative algorithm approaches provable optimality or guarantees failure. Human culture, on this account, is saturated with Institutionalized Stupidity: rule systems that yield worse-than-random outcomes entrenched by tradition, authority, or misplaced confidence. The reductive Anthropomorphism that confines intelligence to Homo sapiens and collapses it into a single scalar metric like IQ is itself an instance of the phenomenon it fails to recognize.