Paper Number
ICIS2025-2433
Paper Type
Complete
Abstract
AI brings its computational advantages for augmenting task performance. However, delegating tasks to AI has received a mixed reception among experts and novices across various task domains. This raises an important question: do experts or novices benefit more from delegating to AI under varying levels of environmental uncertainty? To address these issues, we leverage the agentic IS delegation framework to develop our research model. We conduct an experiment in a portfolio management context. The results show that task-related expertise facilitates effective AI delegation in uncertain task environments but hinders effective AI delegation in low-uncertainty environments. Our findings highlight that, compared to task-related experts, novices benefit more from delegating to AI in low-uncertainty environments and can experience greater losses due to ineffective delegation in uncertain environments, whereas experts exhibit the opposite pattern. Our results contribute to the human-AI delegation framework and offer implications for firms and AI developers.
Recommended Citation
Zhao, Renzhi (Fred) and Gao, Hongyu, "Experts, Novices, and AI: Delegation Decisions in Uncertain Environments" (2025). ICIS 2025 Proceedings. 14.
https://aisel.aisnet.org/icis2025/general_topic/general_topic/14
Experts, Novices, and AI: Delegation Decisions in Uncertain Environments
AI brings its computational advantages for augmenting task performance. However, delegating tasks to AI has received a mixed reception among experts and novices across various task domains. This raises an important question: do experts or novices benefit more from delegating to AI under varying levels of environmental uncertainty? To address these issues, we leverage the agentic IS delegation framework to develop our research model. We conduct an experiment in a portfolio management context. The results show that task-related expertise facilitates effective AI delegation in uncertain task environments but hinders effective AI delegation in low-uncertainty environments. Our findings highlight that, compared to task-related experts, novices benefit more from delegating to AI in low-uncertainty environments and can experience greater losses due to ineffective delegation in uncertain environments, whereas experts exhibit the opposite pattern. Our results contribute to the human-AI delegation framework and offer implications for firms and AI developers.
When commenting on articles, please be friendly, welcoming, respectful and abide by the AIS eLibrary Discussion Thread Code of Conduct posted here.
Comments
02-GeneralTopics