x
CFAR should explicitly focus on AI safety - History — LessWrong