Losing access to an AI tool makes workers less productive than they were before they ever started using the technology in the first place.
Economists typically view AI as a simple plus-one to human effort that improves our baseline. This six-axis loss model shows that we suffer from asymmetric regression when the tool is taken away. Our skills atrophy as we delegate tasks to the machine, making us worse at our jobs when forced to work alone. This creates a dangerous dependency where a simple service disruption can paralyze an entire industry. We are not just getting faster, we are fundamentally changing our own cognitive architecture to rely on external support. The long-term risk of AI is not just replacement, but the permanent degradation of independent human capability.
Dependency Exposure: A Six-Axis Loss Model for AI Service Disruption
SSRN · 6723799
When AI dependency is severed, what is lost? Existing research bifurcates into upstream analyses of why dependency forms and downstream calculations of economic downtime costs, leaving the multi-dimensional structure of losses exposed at severance largely undescribed. This paper introduces the concept of dependency exposure and proposes a Six-Axis Loss Model that describes losses across six structurally independent axes: economic, psychological, cognitive, ontological, emergentive, and sovereign