What Do LLMs Want?

Thomas Cook, Sophia Kazinnik, Zach Modig and Nathan Palmer

No RWP 25-19, Research Working Paper from Federal Reserve Bank of Kansas City

Abstract: Large language models (LLMs) are now used for economic reasoning, but their implicit "preferences” are poorly understood. We study LLM preferences as revealed by their choices in simple allocation games and a job-search setting. Most models favor equal splits in dictator-style allocation games, consistent with inequality aversion. Structural estimates recover Fehr–Schmidt parameters that indicate inequality aversion is stronger than in similar experiments with human participants. However, we find these preferences are malleable: reframing (e.g., masking social context) and learned control vectors shift choices toward payoff-maximizing behavior, while personas move them less effectively. We then turn to a more complex economic scenario. Extending a McCall job search environment, we also recover effective discounting from accept/reject policies, but observe that model responses may not always be rationalizable, and in some cases suggest inconsistent preferences. Efforts to steer LLM responses in the McCall scenario are also less consistent. Together, our results suggest (i) LLMs exhibit latent preferences that may not perfectly align with typical human preferences and (ii) LLMs can be steered toward desired preferences, though this is more difficult with complex economic tasks.

Keywords: large language models; Simulation modeling (search for similar items in EconPapers)
JEL-codes: C61 C63 C68 D14 D83 D91 E20 E21 (search for similar items in EconPapers)
Pages: 62
Date: 2025-11-25
New Economics Papers: this item is included in nep-ain, nep-cbe and nep-exp
References: Add references at CitEc
Citations:

Downloads: (external link)
https://www.kansascityfed.org/research/research-wo ... s/what-do-llms-want/ Full text (text/html)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:fip:fedkrw:102166

Ordering information: This working paper can be ordered from

DOI: 10.18651/RWP2025-19

Access Statistics for this paper

More papers in Research Working Paper from Federal Reserve Bank of Kansas City Contact information at EDIRC.
Bibliographic data for series maintained by Kira Lillard ().