IDEAS home Printed from https://ideas.repec.org/p/arx/papers/2406.05408.html
   My bibliography  Save this paper

Human Learning about AI Performance

Author

Listed:
  • Bnaya Dreyfuss
  • Raphael Raux

Abstract

How do humans assess the performance of Artificial Intelligence (AI) across different tasks? AI has been noted for its surprising ability to accomplish very complex tasks while failing seemingly trivial ones. We show that humans engage in ``performance anthropomorphism'' when assessing AI capabilities: they project onto AI the ability model that they use to assess humans. In this model, observing an agent fail an easy task is highly diagnostic of a low ability, making them unlikely to succeed at any harder task. Conversely, a success on a hard task makes successes on any easier task likely. We experimentally show that humans project this model onto AI. Both prior beliefs and belief updating about AI performance on standardized math questions appear consistent with the human ability model. This contrasts with actual AI performance, which is uncorrelated with human difficulty in our context, and makes such beliefs misspecified. Embedding our framework into an adoption model, we show that patterns of under- and over-adoption can be sustained in an equilibrium with anthropomorphic beliefs.

Suggested Citation

  • Bnaya Dreyfuss & Raphael Raux, 2024. "Human Learning about AI Performance," Papers 2406.05408, arXiv.org.
  • Handle: RePEc:arx:papers:2406.05408
    as

    Download full text from publisher

    File URL: http://arxiv.org/pdf/2406.05408
    File Function: Latest version
    Download Restriction: no
    ---><---

    References listed on IDEAS

    as
    1. George Loewenstein & Ted O'Donoghue & Matthew Rabin, 2003. "Projection Bias in Predicting Future Utility," The Quarterly Journal of Economics, President and Fellows of Harvard College, vol. 118(4), pages 1209-1248.
    2. Sandro Ambuehl & B. Douglas Bernheim & Axel Ockenfels, 2021. "What Motivates Paternalism? An Experimental Study," American Economic Review, American Economic Association, vol. 111(3), pages 787-830, March.
    3. David Danz & Lise Vesterlund & Alistair J. Wilson, 2020. "Belief Elicitation: Limiting Truth Telling with Information on Incentives," NBER Working Papers 27327, National Bureau of Economic Research, Inc.
    4. Viswanath Venkatesh & Fred D. Davis, 2000. "A Theoretical Extension of the Technology Acceptance Model: Four Longitudinal Field Studies," Management Science, INFORMS, vol. 46(2), pages 186-204, February.
    5. Erkal, Nisvan & Gangadharan, Lata & Koh, Boon Han, 2020. "Replication: Belief elicitation with quadratic and binarized scoring rules," Journal of Economic Psychology, Elsevier, vol. 81(C).
    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. Tristan Gagnon-Bartsch & Antonio Rosato, 2024. "Quality Is in the Eye of the Beholder: Taste Projection in Markets with Observational Learning," American Economic Review, American Economic Association, vol. 114(11), pages 3746-3787, November.
    2. Charness, Gary & Gneezy, Uri & Rasocha, Vlastimil, 2021. "Experimental methods: Eliciting beliefs," Journal of Economic Behavior & Organization, Elsevier, vol. 189(C), pages 234-256.
    3. Lata Gangadharan & Philip J. Grossman & Nina Xue, 2021. "Identifying self-image concerns from motivated beliefs: Does it matter how and whom you ask?," Monash Economics Working Papers 2021-17, Monash University, Department of Economics.
    4. Burdea, Valeria & Woon, Jonathan, 2022. "Online belief elicitation methods," Journal of Economic Psychology, Elsevier, vol. 90(C).
    5. Saeideh Sharifi fard & Ezhar Tamam & Md Salleh Hj Hassan & Moniza Waheed & Zeinab Zaremohzzabieh, 2016. "Factors affecting Malaysian university students’ purchase intention in social networking sites," Cogent Business & Management, Taylor & Francis Journals, vol. 3(1), pages 1182612-118, December.
    6. Andrzej Baniak & Peter Grajzl, 2017. "Optimal Liability when Consumers Mispredict Product Usage," American Law and Economics Review, American Law and Economics Association, vol. 19(1), pages 202-243.
    7. Thunström, Linda & Nordström, Jonas & Shogren, Jason F., 2015. "Certainty and overconfidence in future preferences for food," Journal of Economic Psychology, Elsevier, vol. 51(C), pages 101-113.
    8. Chou, Jui-Sheng & Gusti Ayu Novi Yutami, I, 2014. "Smart meter adoption and deployment strategy for residential buildings in Indonesia," Applied Energy, Elsevier, vol. 128(C), pages 336-349.
    9. Philippe Cohard, 2020. "Information Systems Values: A Study of the Intranet in Three French Higher Education Institutions," Post-Print hal-02987225, HAL.
    10. Humphreys, Brad & Ruseski, Jane & Zhou, Li, 2015. "Physical Activity, Present Bias, and Habit Formation: Theory and Evidence from Longitudinal Data," Working Papers 2015-6, University of Alberta, Department of Economics.
    11. Howard Kunreuther & Erwann Michel-Kerjan, 2015. "Demand for fixed-price multi-year contracts: Experimental evidence from insurance decisions," Journal of Risk and Uncertainty, Springer, vol. 51(2), pages 171-194, October.
    12. Melih Engin & Fatih Gürses, 2019. "Adoption of Hospital Information Systems in Public Hospitals in Turkey: An Analysis with the Unified Theory of Acceptance and Use of Technology Model," International Journal of Innovation and Technology Management (IJITM), World Scientific Publishing Co. Pte. Ltd., vol. 16(06), pages 1-19, October.
    13. Michel, Christian, 2017. "Market regulation of voluntary add-on contracts," International Journal of Industrial Organization, Elsevier, vol. 54(C), pages 239-268.
    14. Evans, Alecia & Sesmero, Juan, 2022. "Cooperation in Social Dilemmas with Correlated Noisy Payoffs: Theory and Experimental Evidence," 2021 Annual Meeting, August 1-3, Austin, Texas 322804, Agricultural and Applied Economics Association.
    15. Botond Kőszegi & Matthew Rabin, 2006. "A Model of Reference-Dependent Preferences," The Quarterly Journal of Economics, President and Fellows of Harvard College, vol. 121(4), pages 1133-1165.
    16. Morosan, Cristian, 2016. "An empirical examination of U.S. travelers’ intentions to use biometric e-gates in airports," Journal of Air Transport Management, Elsevier, vol. 55(C), pages 120-128.
    17. Tsung Teng Chen, 2012. "The development and empirical study of a literature review aiding system," Scientometrics, Springer;Akadémiai Kiadó, vol. 92(1), pages 105-116, July.
    18. Abdesamad Zouine & Pierre Fenies, 2014. "The Critical Success Factors Of The ERP System Project: A Meta-Analysis Methodology," Post-Print hal-01419785, HAL.
    19. Debora Bettiga & Lucio Lamberti & Emanuele Lettieri, 2020. "Individuals’ adoption of smart technologies for preventive health care: a structural equation modeling approach," Health Care Management Science, Springer, vol. 23(2), pages 203-214, June.
    20. Kertcher, Zack & Venkatraman, Rohan & Coslor, Erica, 2020. "Pleasingly parallel: Early cross-disciplinary work for innovation diffusion across boundaries in grid computing," Journal of Business Research, Elsevier, vol. 116(C), pages 581-594.

    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:2406.05408. 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.