x
AI-Assisted Alignment - History — LessWrong