Robin Edgard Ulrik Mertens
To solve the problem of meaning degradation when AI accelerates decision-making, a closed-loop architecture is proposed that preserves human interpretive control while supporting AI analysis.
In the process of scaling decision-making with AI, intent is translated into criteria, metrics, and allocation rules, causing meaning to drift and become untraceable.
A closed-loop architecture is introduced to constrain how meaning is translated into decisions. The 'Operating Spine' serves as a structural unit of analysis to trace the connection between intent and action across governance layers.
It demonstrates that in public institutions, capital allocation, and AI-mediated environments, the durability of decision rules determines long-term institutional reliability, making drift and coherence directly observable within decision systems.