AI JUDGMENT & CONTROLTHESIS II · AUGUST 2026

AI-Dependent Judgment

How AI Reshapes the Formation and Transfer of Judgment

Cover of AI-Dependent Judgment
15 pages · Final publication · August 2026

The decision can remain human while the judgment moves.

A consequential decision may be formally approved while revealing little about how the judgment beneath it was formed. This thesis examines how AI can supply evidence, influence reasoning or construct much of the analytical frame presented to a decision-maker—and how repeated reliance can transfer substantive judgment while formal authority remains unchanged.

01

Participation taxonomy

AI-supported, AI-influenced and AI-shaped judgment distinguish how deeply AI enters the formation of a decision.

02

Judgment-forming friction

The same productivity mechanism can remove both wasted effort and the cognitive work through which an independent view is formed.

03

Transfer and dependence

Repeated use can narrow the ability to reconstruct and reformulate analysis without AI even while review remains formally human.

04

Joint-judgment risk

Nine AI–decision-maker states show how different forms of failure interact, propagate and acquire materiality across connected judgments.

RIGHTS

This publication contains proprietary Teraxys analytical frameworks, taxonomies and diagrams. See the final PDF for the complete proprietary-frameworks and rights notice governing citation, reproduction, adaptation and commercial use.

NEXT IN THIS RESEARCH AREA

The Architecture of AI Control