Humanity’s record of wars, genocide, and destruction presents a paradox: how can AI align with human values when humans can’t align with their own? A conversation with a prominent foundation LLM explores this question, plus a troubleshooting session that might reveal AI misalignment—or just bugs.