Cognitive uncontainability - History — AI Alignment Forum