Can AI Agents with Divergent Interests Learn to Prevent Civilizational Failures?lesswrong.com·2 pts·abranti·0