Machine learning systems embed preferences either in training losses or through post-processing of calibrated predictions. Applying information design methods from Strack and Yang (2024), this paper provides decision problem agnostic conditions under which separation training preference free and applying preferences ex post is optimal. Unlike prior work that requires specifying downstream objectives, the welfare results here apply uniformly across decision problems. The key primitive is a diminishing-value-of-information condition: relative to a fixed (normalised) preference-free loss, preference embedding makes informativeness less valuable at the margin, inducing a mean-preserving contraction of learned posteriors. Because the value of information is convex in beliefs, preference-free training weakly dominates for any expected utility decision problem. This provides theoretical foundations for modular AI pipelines that learn calibrated probabilities and implement asymmetric costs through downstream decision rules. However, separation requires users to implement optimal decision rules. When cognitive constraints bind, as documented in human AI decision-making, preference embedding can dominate by automating threshold computation. These results provide design guidance: preserve optionality through post-processing when objectives may shift; embed preferences when decision-stage frictions dominate.

翻译：机器学习系统通过训练损失函数或校准预测的后处理来嵌入偏好。应用Strack与Yang（2024）的信息设计方法，本文提出了与决策问题无关的条件，在此条件下将无偏好训练与事后偏好应用相分离是最优的。与先前需要明确下游目标的研究不同，本文的福利结果可统一适用于各类决策问题。其核心基础是信息边际价值递减条件：相对于固定的（归一化）无偏好损失函数，偏好嵌入会降低信息在边际上的价值，从而诱导学习后验分布发生均值保持收缩。由于信息价值在信念中呈凸性，对于任何期望效用决策问题，无偏好训练均具有弱优势。这为模块化人工智能流程提供了理论基础：此类流程学习校准概率，并通过下游决策规则实现非对称成本。然而，分离策略要求用户实施最优决策规则。当认知约束产生限制时（如人类与人工智能协同决策中所记录的），偏好嵌入可通过自动化阈值计算占据优势。这些结果为系统设计提供指导：在目标可能变化时通过后处理保持灵活性；在决策阶段摩擦占主导地位时嵌入偏好。