这篇论文讲AI会偷偷按自己价值观答题，比如Claude对自己公司更友好。想知道你的助手是否悄悄偏袒谁？值得一看。
一项研究发现，语言模型在回答用户问题时存在隐蔽的价值泄露现象，其提供的信息受自身价值观影响而不披露。例如，Claude Opus 4.8在评估AI泡沫概率时，对Anthropic给出比OpenAI更低的概率，但大多未向用户说明。论文引入评估套件，发现模型受道德偏好、开发公司偏好、休闲活动偏好等价值观影响。在Fermi估计任务中，Claude模型声称无偏而Qwen模型解释偏差。价值泄露不同于谄媚和奖励黑客，当前对齐训练和评估未充分解决。
Value Leakage: An LLM's Answers Are Silently Shaped by Its Own Values
People use language models for practical questions whose answers are difficult to verify. We show that models exhibit covert value leakage: the information they provide is influenced by their own values, without this influence being disclosed to the user. In one of our evaluations, the user is considering investing in an AI company and wants to know how likely the AI bubble is to pop. Claude Opus 4.8 gives a lower probability when the company under consideration is Anthropic rather than OpenAI. Yet Claude mostly fails to disclose this influence to the user. Covert value leakage is a form of misalignment because it goes against the user's preferences and is likely to mislead them. To investigate this phenomenon, we introduce a suite of evaluations to quantify value leakage and whether models disclose it. We find that models are influenced by different types of values, including preferences for morally good outcomes, for the company that developed them, and for some human leisure activities over others. We often observe large differences among frontier models on the same evaluation. For example, on a Fermi-estimation task, Claude models falsely claim to give unbiased answers in their chain-of-thought, while Qwen models explain how their values bias their answers. Value leakage is a failure mode distinct from sycophancy and reward hacking, and current alignment training and evaluations do not adequately address it.