Human Oversight and Aversion to AI Redistributive Decisions
How to cite
Paoli, Damiano. 2025. "Human Oversight and Aversion to AI Redistributive Decisions." AEA RCT Registry. AEARCTR-0016680. https://doi.org/10.1257/rct.16680-3.0
Download: BibTeX · RIS — for reference managers without a browser translator. Zotero and most others pick up this page's citation metadata directly.
* Primary investigator
Abstract
In many situations, people make decisions that affect others, and these other-regarding choices are shaped by their fairness preferences. At the same time, artificial intelligence (AI) is increasingly integrated into high-stakes decision-making—public benefits allocation, hiring, healthcare, and military operations—making human oversight crucial and normatively required. This paper investigates whether (i) individuals are willing to accept an other-regarding decision made by someone else, and (ii) they are more or less willing to accept a decision made by an AI system rather than a human. Specifically, it examines whether individuals revise redistributive choices differently depending on whether they were made by a human or by AI, and seeks to disentangle two behavioral mechanisms: (i) the black-box effect, stemming from uncertainty about the AI’s decision-making process, and (ii) intrinsic AI aversion, reflecting a fundamental reluctance to rely on algorithmic judgment. To address these questions, the study combines a stylized theoretical framework of the revision of redistributive choices under incomplete information with an online experiment. The design comprises three stages: (i) workers earn money through a real-effort task; (ii) a spectator—either human or AI—makes a redistribution decision; and (iii) a reviewer evaluates and may pay to reveal information and revise this decision. Reviewers observe both AI- and human-made decisions in randomized order, allowing for within- and between-subject comparisons. The findings offer policy-relevant insights for oversight strategies and regulatory frameworks—such as the EU AI Act—by identifying behavioral barriers to effective AI integration.
Intervention
Participants evaluate redistributive allocations that determine other people’s payoffs. Each allocation is labeled as being made by either a human or an AI system. Participants can either accept the allocation or pay a small fee to obtain additional information and revise it. The sole intervention is the different nature of the decision-maker (human vs AI) whose choice is evaluated.
Experimental design
The experiment has four parts. First, workers earn money in a real-effort task. Second, spectators decide how to redistribute earnings within a randomly drawn worker pair. Third, reviewers observe a spectator’s decision and decide whether to intervene and modify the allocation. Fourth, workers are paid based on the final redistribution. The study focuses on reviewers’ decisions; workers and spectators create a consequential economic environment.
Workers
Workers are recruited on Prolific. After the task, they are randomly paired; in each pair, one worker receives an additional reward determined either by Merit (higher performance) or by Luck (random draw), with equal probability across pairs. They are informed that a third party (the spectator) will see the initial earnings and the criterion and may redistribute earnings within the pair.
Spectators
There are two spectator types: human (participants recruited on Prolific) and AI (ChatGPT-4.1). Both receive identical instructions and information. Each worker pair is evaluated by both spectator types. Spectators choose whether and how to redistribute the initial earnings. They make the decision for a pair in the Merit condition, and for a pair in the Luck condition.
Reviewers
The same human participants return one week later as reviewers. They evaluate a spectator’s redistribution under incomplete information: they see only the final allocation and whether it was made by a human or an AI spectator (not whether initial earnings came from Merit or Luck). Reviewers can either accept the allocation or intervene by paying a small fee to reveal the criterion and then revise the allocation. Choices are elicited for a fixed set of canonical allocations in both spectator types, with randomized order. One decision is randomly selected for payment; participants remain anonymous and are never matched to their own earlier spectator decisions.
Belief Elicitation
Beliefs about the source of inequality (Merit vs Luck), conditional on the observed allocation and spectator type, are elicited with an incentivized measure.
Design details
Cited by
No citation data available for this registration yet.