x
AI Alignment and Recognition — LessWrong