IDEAS home Printed from https://ideas.repec.org/p/arx/papers/2603.29317.html

Should I State or Should I Show? Aligning AI with Human Preferences

Author

Listed:
  • Keaton Ellis
  • Wanying Huang

Abstract

As AI agents become more autonomous, properly aligning their objectives with human preferences becomes increasingly important. We study how effectively an AI agent learns a human principal's preference in choice under risk via stated versus revealed preferences. We conduct an online experiment in which subjects state their preferences through written instructions ("prompts") and reveal them through choices in a series of binary lottery questions ("data"). We find that on average, an AI agent given revealed-preference data predicts subjects' choices more accurately than an AI agent given stated-preference prompts. Further analysis suggests that the gap is driven by subjects' difficulty in translating their own preferences into written instructions. When given a choice between which information source to give to an AI agent, a large portion of subjects fail to select the more informative one. Moreover, when predictions from the two sources conflict, we find that the AI agent aligns more frequently with the prompt, despite its lower accuracy. Overall, these results highlight the revealed preference approach as a powerful mechanism for communicating human preferences to AI agents, but its success depends on careful implementation.

Suggested Citation

  • Keaton Ellis & Wanying Huang, 2026. "Should I State or Should I Show? Aligning AI with Human Preferences," Papers 2603.29317, arXiv.org, revised May 2026.
    • Handle: RePEc:arx:papers:2603.29317
    as

    Download full text from publisher

    File URL: http://arxiv.org/pdf/2603.29317
    File Function: Latest version
    Download Restriction: no
    ---><---

    References listed on IDEAS

    as
    1. Leonidas Spiliopoulos & Andreas Ortmann, 2018. "The BCD of response time analysis in experimental economics," Experimental Economics, Springer;Economic Science Association, vol. 21(2), pages 383-433, June.
    Full references (including those not matched with items on IDEAS)

    Most related items

    These are the items that most often cite the same works as this one and are cited by the same works as this one.
    1. Strittmatter, Anthony & Sunde, Uwe & Zegners, Dainis, 2022. "Speed, Quality, and the Optimal Timing of Complex Decisions: Field Evidence," Rationality and Competition Discussion Paper Series 317, CRC TRR 190 Rationality and Competition.
    2. Syngjoo Choi & Jeongbin Kim & Eungik Lee & Jungmin Lee, 2022. "Probability Weighting and Cognitive Ability," Management Science, INFORMS, vol. 68(7), pages 5201-5215, July.
    3. Markus Jung & Mischa Seiter, 2021. "Towards a better understanding on mitigating algorithm aversion in forecasting: an experimental study," Journal of Management Control: Zeitschrift für Planung und Unternehmenssteuerung, Springer, vol. 32(4), pages 495-516, December.
    4. Backhaus, Teresa & Huck, Steffen & Leutgeb, Johannes & Oprea, Ryan, 2023. "Learning through period and physical time," Games and Economic Behavior, Elsevier, vol. 141(C), pages 21-29.
    5. Aurélien Baillon & Zhenxing Huang & Asli Selim & Peter P. Wakker, 2018. "Measuring Ambiguity Attitudes for All (Natural) Events," Econometrica, Econometric Society, vol. 86(5), pages 1839-1858, September.
    6. Sean Duffy & John Smith, 2025. "An economist and a psychologist form a line: What can imperfect perception of length tell us about stochastic choice?," Theory and Decision, Springer, vol. 99(3), pages 701-734, November.
    7. Fraser, Iain & Balcombe, Kelvin & Williams, Louis & McSorley, Eugene, 2021. "Preference stability in discrete choice experiments. Some evidence using eye-tracking," Journal of Behavioral and Experimental Economics (formerly The Journal of Socio-Economics), Elsevier, vol. 94(C).
    8. Cary Frydman & Ian Krajbich, 2022. "Using Response Times to Infer Others’ Private Information: An Application to Information Cascades," Management Science, INFORMS, vol. 68(4), pages 2970-2986, April.
    9. Deck, Cary & Jahedi, Salar & Sheremeta, Roman, 2021. "On the consistency of cognitive load," European Economic Review, Elsevier, vol. 134(C).
    10. Carlos Alós-Ferrer & Ernst Fehr & Nick Netzer, 2021. "Time Will Tell: Recovering Preferences When Choices Are Noisy," Journal of Political Economy, University of Chicago Press, vol. 129(6), pages 1828-1877.
    11. Banerjee, Priyodorshi & Das, Tanmoy, 2021. "Risky decision under laboratory deadline with experience and indirect self-selection," Journal of Behavioral and Experimental Finance, Elsevier, vol. 29(C).
    12. Carlos Alós-Ferrer & Johannes Buckenmaier, 2021. "Cognitive sophistication and deliberation times," Experimental Economics, Springer;Economic Science Association, vol. 24(2), pages 558-592, June.
    13. Anna Conte & Gianmarco Santis & John D. Hey & Ivan Soraperra, 2023. "The determinants of decision time in an ambiguous context," Journal of Risk and Uncertainty, Springer, vol. 67(3), pages 271-297, December.
    14. Florian Diekert & Kjell Arne Brekke, 2022. "Groups discipline resource use under scarcity," Theory and Decision, Springer, vol. 92(1), pages 75-103, February.
    15. Chapkovski, Philipp & Zihlmann, Christian, 2019. "Introducing otree_tools: A powerful package to provide process data for attention, multitasking behavior and effort through tracking focus," Journal of Behavioral and Experimental Finance, Elsevier, vol. 23(C), pages 75-83.
    16. Buffat, Justin & Praxmarer, Matthias & Sutter, Matthias, 2023. "The intrinsic value of decision rights: A replication and an extension to team decision making," Journal of Economic Behavior & Organization, Elsevier, vol. 209(C), pages 560-571.
    17. Avoyan, Ala & He, Haoran & Lu, Kelin, 2025. "Team work under time constraints," Journal of Economic Behavior & Organization, Elsevier, vol. 237(C).
    18. Hikaru Kawarazaki & Minhaj Mahmud & Yasuyuki Sawada & Mai Seki, 2023. "Haste Makes No Waste: Positive Peer Effects of Classroom Speed Competition on Learning," Oxford Bulletin of Economics and Statistics, Department of Economics, University of Oxford, vol. 85(4), pages 755-772, August.
    19. Haji, Anouar El & Krawczyk, Michał & Sylwestrzak, Marta & Zawojska, Ewa, 2019. "Time pressure and risk taking in auctions: A field experiment," Journal of Behavioral and Experimental Economics (formerly The Journal of Socio-Economics), Elsevier, vol. 78(C), pages 68-79.
    20. Schütze, Tobias & Spitzer, Carsten & Wichardt, Philipp C., 2025. "Nudging: An experiment on transparency, accounting for reactance and response time," Journal of Economic Psychology, Elsevier, vol. 107(C).

    More about this item

    NEP fields

    This paper has been announced in the following NEP Reports:

    Statistics

    Access and download statistics

    Corrections

    All material on this site has been provided by the respective publishers and authors. You can help correct errors and omissions. When requesting a correction, please mention this item's handle: RePEc:arx:papers:2603.29317. See general information about how to correct material in RePEc.

    If you have authored this item and are not yet registered with RePEc, we encourage you to do it here. This allows to link your profile to this item. It also allows you to accept potential citations to this item that we are uncertain about.

    If CitEc recognized a bibliographic reference but did not link an item in RePEc to it, you can help with this form .

    If you know of missing items citing this one, you can help us creating those links by adding the relevant references in the same way as above, for each refering item. If you are a registered author of this item, you may also want to check the "citations" tab in your RePEc Author Service profile, as there may be some citations waiting for confirmation.

    For technical questions regarding this item, or to correct its authors, title, abstract, bibliographic or download information, contact: arXiv administrators (email available below). General contact details of provider: http://arxiv.org/ .

    Please note that corrections may take a couple of weeks to filter through the various RePEc services.

    IDEAS is a RePEc service. RePEc uses bibliographic data supplied by the respective publishers.