How Ensembling AI and Public Managers Improves Decision-Making

Florian Keppeler, Jana Borchert, Mogens Jin Pedersen and Vibeke Lehmann Nielsen

No 2yf6r, OSF Preprints from Center for Open Science

Abstract: Artificial Intelligence (AI) applications are transforming public sector decision-making. However, most research conceptualizes AI as a form of specialized algorithmic decision support tool. In contrast, this study introduces the concept of human-AI ensembles, where humans and AI tackle the same tasks together, rather than specializing in certain parts. We argue that this is particularly relevant for many public sector decisions, where neither human nor AI-based decision-making has a clear advantage over the other in terms of legitimacy, efficacy, or legality. We illustrate this design theory within access to public employment, focusing on two key areas: (a) the potential of ensembling human and AI to reduce biases and (b) the inclinations of public managers to use AI advice. Study 1 presents evidence from the assessment of real-life job candidates (n = 2,000) at the intersection of gender and ethnicity by public managers compared to AI. The results indicate that ensembled decision- making may alleviate ethnic biases. Study 2 examines how receptive public managers are to AI advice. Results from a pre-registered survey experiment involving managers (n = 538 with 4 observations each) show that decision-makers, when reminded of the unlawfulness of hiring discrimination, prioritize AI advice over human advice.

Date: 2024-08-21
New Economics Papers: this item is included in nep-ain, nep-exp and nep-inv
References: View references in EconPapers View complete reference list from CitEc
Citations:

Downloads: (external link)
https://osf.io/download/66c5c875d696426ac4c5d346/

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:osf:osfxxx:2yf6r

DOI: 10.31219/osf.io/2yf6r

Access Statistics for this paper

More papers in OSF Preprints from Center for Open Science
Bibliographic data for series maintained by OSF (contact@cos.io).