Educational Skill Building versus AI Prosthetic Dependency

A productive distinction exists between technology as cognitive enhancement and technology as functional prosthetic. An enhancement amplifies a capacity that has already been internalized — when the tool is removed, the underlying competence persists and may even be stronger for having been exercised under load. A prosthetic, by contrast, substitutes for a function that was never developed or has atrophied. The difference is not in the tool itself but in the developmental sequence in which it is introduced.

The GPS case makes this concrete. A navigator who has built genuine wayfinding skill through dead reckoning, map reading, and environmental cue interpretation, and who then adopts GPS, retains full capability when the technology fails. Someone who has only ever navigated by GPS has acquired infrastructure dependency, not skill. Crucially, the prosthetic user may be indistinguishable from the skilled user in normal operating conditions — the deficit only becomes visible under stress or constraint.

This framework reframes the debate about AI in education. The output equivalence problem — the fact that AI-assisted and independently produced work can appear identical — obscures a more important divergence in what has actually been acquired. The pedagogically significant question is not whether AI should be permitted but at what stage of skill formation its introduction forecloses the development of the deeper competency it is meant to support. This requires a developmental theory of learning that is prior to and independent of any particular technology — a form of wisdom about human cognitive growth that cannot be derived from the tools themselves.