Definition
The ratio of AI-predicted civilian and infrastructure impact to actual post-strike assessed impact, tracked as a running calibration.
Formula
CDEA_running = median(CDE_predicted / CDE_actual)
Proposed formula, pending validation pilots.
At a Glance
Category
AI Calibration & Trust Integrity
Decision Supported
Is the CDE model giving commanders and JAG reliable risk estimates, or are legal compliance decisions based on systematically flawed predictions?