Understanding Human-AI Trust in Military AI Applications: A Multistage Human Factors Framework

Abstract

Artificial intelligence (AI) is increasingly used in military systems to support intelligence, surveillance, reconnaissance, autonomous operations, logistics, command and control, and decision support. Effective use, however, depends not only on technical performance but also on how operators perceive, interpret, evaluate, and rely on AI outputs under uncertainty, time pressure, and adversarial interference. This study presents a conceptual integrative review informed by a structured supplementary literature search and synthesis. It synthesizes human factors, human–automation trust, human–AI teaming, military AI, and industrial engineering literature to develop an iterative multistage framework of human–AI trust in military applications. Trust is conceptualized as dynamic situational trust in a specific military AI system during task interaction, while dispositional, initial, and learned trust are treated as antecedent conditions. The framework distinguishes four analytically separable but functionally interdependent stages—perception, interpretation, evaluation, and trust judgment and reliance decision—and identifies four military-relevant moderators: system transparency, explainability, and behavioral predictability; data quality and information uncertainty; adversarial vulnerability; and operational and organizational context. The framework clarifies conceptual pathways to calibrated reliance, overtrust, and undertrust, and derives implications for task analysis, interface evaluation, human-centered design, operator training, and empirical validation.