Generative AI in Managerial Decision-Making: A Systematic Literature Review of Trust, AI Literacy, and Decision Quality

Abstract

Generative artificial intelligence (GenAI) is rapidly entering managerial decision processes, yet evidence remains fragmented regarding when AI-generated advice should be trusted, how managers should evaluate it, and how decision quality should be assessed. This study systematically reviews GenAI-supported managerial decision-making by integrating PRISMA 2020 screening, bibliometric mapping, author-keyword co-occurrence analysis, thematic synthesis, and the Theory–Context–Characteristics–Methodology (TCCM) framework. A Scopus export containing 3,500 records was screened using temporal, language, document-type, topical, and contextual criteria, yielding 99 studies published between 2024 and 2026. The final corpus comprises 342 unique authors across 94 publication sources. Four themes dominate: decision augmentation and analytics; trust, governance, and explainability; human–AI collaboration and capability; and decision quality and organizational outcomes. Trust is more often represented through governance, transparency, explainability, calibration, and human oversight than as an explicitly measured construct, while AI literacy appears explicitly in only one study. Decision quality is likewise rarely conceptualized directly and is commonly proxied by accuracy, efficiency, or predictive performance. The review proposes an integrative model in which GenAI decision-support capabilities influence managerial decision quality through calibrated trust and human–AI collaboration, with AI literacy functioning as a critical capability for verification and appropriate reliance. The study concludes with a TCCM-based research agenda for human-centered GenAI decision research.   Keywords: generative AI; managerial decision-making; trust in AI; AI literacy; decision quality