Existential risk - History — AI Alignment Forum