The Limits of AI-Driven Allocation: Optimal Screening under Aleatoric Uncertainty

2026-05-08

AI Summary

This study addresses the vulnerability of AI-driven resource allocation to aleatoric uncertainty inherent in individual risk assessments, which can lead to misallocation. The authors propose a two-stage framework: first conducting ground-truth screening for a subset of individuals, then performing targeted allocation under a fixed budget. Theoretical analysis reveals that the optimal strategy screens individuals near the algorithmic assignment boundary while directly covering the highest-risk group. Depending on the level of uncertainty, screening and algorithmic targeting act either as complements or substitutes. Leveraging probabilistic modeling and optimization theory, the framework is empirically validated using real-world data from social protection and humanitarian demining programs in Colombia, demonstrating significantly improved allocation efficiency under high uncertainty.

Abstract

The rise of machine learning has shifted targeted resource allocation in policy and humanitarian settings toward algorithmic targeting based on predicted risk scores. This approach is typically cheaper and faster than traditional screening procedures that directly observe the latent vulnerability status through physical verification. Yet, even access to the true conditional vulnerability probability cannot eliminate misallocation: aleatoric uncertainty over individual vulnerability status is irreducible, and probabilistic targeting inevitably misallocates some resources. In this work we study how screening and algorithmic targeting should be optimally combined in a two-stage allocation framework where a screening stage observes true outcomes for a subset of units before a final allocation stage assigns the resource under a fixed coverage budget. We show that the optimal strategy screens units at the margin of algorithmic allocation, while directly targeting the highest-risk units. Furthermore, we empirically characterize when screening and algorithmic targeting act as complements or substitutes: efficiency gains from screening grow as the aleatoric uncertainty in the population increases. We illustrate our framework with applications in income-based social protection programs and humanitarian demining in Colombia, where the tension between screening costs and allocation efficiency is operationally consequential.

Problem

• aleatoric uncertainty
• resource allocation
• algorithmic targeting
• screening
• misallocation

Innovation

• aleatoric uncertainty
• algorithmic targeting
• optimal screening
• two-stage allocation
• resource allocation