This toy model uses Engineering Quality as a state Q(t) (automation, CI speed, tests,
guardrails, refactoring discipline). Quality acts as an inverse multiplier on operations and maintenance cost.
AI primarily amplifies the control loop (change capability), accelerating software volume growth (V(t)).
This chart compares total cost for three regimes (Average, Quality Investment, Sloppiness), each shown
with and without AI. AI-on uses aiAmplification = 2.0; AI-off uses
aiAmplification = 1.0.
C = C_D + C_O + C_M with
C_O ∝ λ·(1/Q)^b, C_M ∝ B·(1/Q)^g, and
λ ∝ V·r·(1/Q)^a·g(φ).