TY  - RPRT
        AU  - Haq, Omar Abdel
AU  - Chandra, Amitabh
AU  - Jagelka, Tomáš
AU  - Luttmer, Erzo F.P.
AU  - Schwartzstein, Joshua
        TI  - Revealing Life Preferences Through LLMs
        PY  - 2026/May/
        PB  - Institute of Labor Economics (IZA)
        CY  - Bonn
        T2  - IZA Discussion Paper
        IS  - 18634
        UR  - https://www.iza.org/publications/dp18634
        AB  - Large Language Models (LLMs) are trained on a prodigious corpus of human writing and may reveal human preferences over characteristics of life courses, such as income, longevity, and working conditions. We present OpenAI's GPT-5.4 and a broadly representative sample of Americans with pairs of life stories and ask them to choose the life they would prefer for themselves. A person's choice is better predicted by the LLM's choice than by another person’s choice over the same stories, and LLM valuations of several life attributes are similar to those derived from human responses. Our results suggest that LLM responses offer a scalable and cost-effective complement to existing methods for studying human preferences.
        KW  - generative AI
KW  - preference estimation methods
KW  - choice experiments
KW  - survey validation
        ER  -