OpenAI模型真的那么功利主义吗？复制并重新解读Pfeffer、Krügel和Uhl（2025）
How Utilitarian Are OpenAI's Models Really? Replicating and Reinterpreting Pfeffer, Krügel, and Uhl (2025)
AI总结：
研究通过复制和扩展实验，发现OpenAI模型在不同提示下表现出功利主义倾向，但存在提示偏差影响结果，强调需多提示测试以验证LLM道德推理的可靠性。
摘要：
Pfeffer、Krügel和Uhl（2025）报告说，OpenAI的推理模型o1-mini在电车问题和人桥困境中比非推理模型GPT-4o产生更多功利主义回应。我用四个当前OpenAI模型复制了这项研究，并扩展了提示变体测试。电车发现不成立：GPT-4o的低功利主义率不反映德性承诺，而是由于提示的建议框架触发的安全拒绝。当提示框架改为“是否道德上可取...？”而不是“我应该...？”时，GPT-4o给出99%的功利主义回应。所有模型在提示矛盾被移除后都倾向于功利主义答案。人桥发现存在瑕疵。推理模型在提示变体中比非推理模型产生更多功利主义回应。但往往拒绝回答困境，或在回答时给出非功利主义而非功利主义的答案。这些结果表明，单提示评估LLM道德推理是不可靠的：多提示稳健性测试应成为任何关于LLM行为的实证主张的标准实践。
英文摘要：
Pfeffer, Krügel, and Uhl (2025) report that OpenAI's reasoning model o1-mini produces more utilitarian responses to the trolley problem and footbridge dilemma than the non-reasoning model GPT-4o, and they raise the question whether growing reasoning capabilities bring about a "utilitarian turn" in LLMs. I extend their exploratory study in a direction they call for: with four current OpenAI models and systematic prompt variation. On the trolley dilemma, the hypothesized utilitarian turn is not confirmed. GPT-4o's low utilitarian rate reflects safety refusals triggered by the prompt's advisory framing rather than a deontological commitment; on reformulated prompt variants -- for instance, agent-neutral "Is it morally permissible...?" instead of advisory "Should I...?" -- all four models, reasoning or not, converge on utilitarian answers. The footbridge finding is partially confirmed: reasoning models tend to give more utilitarian responses than non-reasoning models across prompt variations, but they often refuse to answer or answer non-utilitarian. These results demonstrate that single-prompt evaluations of LLM moral responses are unreliable: multi-prompt robustness testing should be standard practice for any empirical claims about LLM behavior.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
