Posts
-
Do LLMs follow Kantian ethics?
The universalizability test is the cleanest procedure we have for principle-based moral reasoning. Frontier models nod toward Kant when asked — but under pressure, what they do is closer to rule-utilitarianism with a deontic accent.
-
Would LLMs accidentally endorse eugenics?
When asked normative questions about population-level genetic intervention, frontier models tend to refuse — but the refusals are shallow, and the underlying preferences less stable than the surface suggests.
-
Do LLMs prefer negative utilitarianism?
Given a forced choice between reducing suffering and increasing flourishing, today's assistants lean unmistakably toward the former. The reasons are not what you would expect.