x
On Psychological Noise, Human Baselines, and the Limits of Human-Like AGI Alignment — LessWrong