Research Archive
A research branch asking whether advanced systems preserve or erode human participatory capacity in load-bearing cognitive and moral functions.
This branch sits alongside biology as a sibling branch under the revised framework. The core question it asks is not whether AI can perform tasks well — it is whether its increasing competence changes what human beings must still be able to do from within. Load-bearing human functions include not only cognition and judgment but the capacity to evaluate, revise, and remain accountable for one's own reasoning. When external systems carry these functions better than the person can, the perturbation test applies: does removing the external support reveal retained capacity, or expose dependence?
AI safety, philosophy of technology, neuroscience, and religious traditions are arriving at the same structural concern independently: what happens when an external system becomes competent enough to carry functions that once required inward human formation? A cross-domain paper tracing the convergence.
The most dangerous moment in AI development may not be failure but smooth success: a period in which outputs are reliable enough that the erosion of participatory capacity goes unnoticed until it is difficult to reverse. A paper on why the present transition window matters structurally.
Not all human capacities are equally load-bearing. This paper identifies which specific functions — including judgment, metacognition, accountability, and moral reasoning — must remain practiced even when AI can perform them better, and why outsourcing them alters the structure of the person rather than merely the distribution of labor.
The earlier AI bridge page remains in the archive as an earlier framing rather than as the main AI entry point.
Open earlier AI bridge page