Replication Data for: From Faces to Politics: Vision-Language Models (Sometimes) Link Visual Demographic Characteristics to Ideological Labels

When foundation models analyze political content, do they use demographic characteristics as shortcuts for ideological attribution? We conducted detailed experiments with GPT-4o-mini and validated key findings across GPT-4o and LLaVA, using identical, ideologically neutral campaign advertisements with systematically varied candidate demographics. All models consistently attributed more liberal ideologies to women than men. These effects exceeded real-world gender differences from a nationally representative survey. However, racial associations differed by model: strong in GPT-4o-mini (where Black candidates received substantially more liberal attributions), attenuated in GPT-4o, and insignificant in LLaVA. These demographic effects persisted across temperature settings, prompt variations, and even explicit debiasing instructions in GPT-4o-mini. Our findings reveal that visual demographic features can shape AI outputs in ways that vary across models, with implications for applications such as content classification.

当基础模型（foundation model）分析政治类内容时，它们是否会将人口统计学特征作为意识形态归因的快捷方式？我们针对GPT-4o-mini开展了详尽的实验，并在GPT-4o与LLaVA上验证了核心结论，实验所用素材为意识形态中立的竞选广告，且系统地调整了候选人的人口统计学特征。所有模型均一致地将女性判定为更偏向自由主义意识形态的比例高于男性，该效应超过了全国代表性调查中观察到的现实性别差异。不过，种族关联效应因模型而异：在GPT-4o-mini中效应显著（黑人候选人被赋予显著更偏向自由主义的归因），在GPT-4o中效应减弱，而在LLaVA中则无统计学显著性。在GPT-4o-mini中，这些人口统计学特征带来的效应在不同温度参数、不同提示词变体，甚至是显式去偏指令下均保持稳定。本研究结果表明，视觉人口统计学特征能够以因模型而异的方式影响AI输出，这对内容分类等应用场景具有重要启示意义。