CONCEPT
Single-Loop and Double-Loop Learning
Argyris's foundational distinction between changing actions to achieve existing goals and changing the goals themselves — the conceptual lever by which the AI transition becomes legible as a learning event rather than merely a technological one.
Single-loop learning adjusts behavior to better achieve existing objectives; double-loop learning interrogates the objectives themselves. A developer who masters a new framework is learning in single loops. An engineer who reconceives what engineering is in the age of AI is learning in double loops. The distinction, developed by Argyris and Donald Schön across four decades, identifies why most organizational responses to transformative change remain structurally inadequate: they optimize within a value system that the change has rendered obsolete. The AI moment is not a tool adoption problem. It is a governing-variable problem, and the difference determines whether practitioners emerge from the transition with expanded capability or merely with new techniques fastened to old identities.
In The You On AI Field Guide
The distinction originates in Argyris and Schön's 1978 Organizational Learning, built from decades of empirical work with executives, consultants, and professional teams. Single-loop learning is the thermostat that notices the room is