Delegation and Verification Under AI

2026-03-03

AI Summary

This study addresses the tension that arises when AI systems are embedded into institutional workflows, where workers must balance task delegation to AI against the effort invested in verification. Institutional evaluations, typically based on outcomes, often misalign with workers’ private costs, thereby undermining overall output quality. The paper models this as a rational worker’s optimization problem and constructs a tripartite interaction framework among workers, AI, and institutions using game theory and mechanism design. It formally demonstrates—for the first time—that even in the absence of cognitive biases, minor differences in workers’ verification capabilities can trigger a phase transition in delegation behavior, leading to quality inequality: while AI enhances performance for highly reliable workers, it systematically degrades the institutional evaluation of outputs from others.

Technology Category

• Philosophy and Ethics of AI: Bias, Fairness & Equity
• Multiagent Systems: Mechanism Design
• Humans and AI: Other Foundations of Human Computation & AI

Application Category

• Economics, Online Markets and Human Computation: Fairness and ethical considerations in crowd work and in human-in-the-loop AI systems
• Responsible Web: Human-perceived consequences of algorithmic deployment on the web
• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics

Abstract

As AI systems enter institutional workflows, workers must decide whether to delegate task execution to AI and how much effort to invest in verifying AI outputs, while institutions evaluate workers using outcome-based standards that may misalign with workers' private costs. We model delegation and verification as the solution to a rational worker's optimization problem, and define worker quality by evaluating an institution-centered utility (distinct from the worker's objective) at the resulting optimal action. We formally characterize optimal worker workflows and show that AI induces *phase transitions*, where arbitrarily small differences in verification ability lead to sharply different behaviors. As a result, AI can amplify workers with strong verification reliability while degrading institutional worker quality for others who rationally over-delegate and reduce oversight, even when baseline task success improves and no behavioral biases are present. These results identify a structural mechanism by which AI reshapes institutional worker quality and amplifies quality disparities between workers with different verification reliability.

Problem

• AI delegation
• verification
• institutional evaluation
• worker quality
• outcome-based standards

Innovation

• delegation
• verification
• phase transitions
• institutional worker quality
• AI-human collaboration