Human-Ai Collaboration in Management: Opportunities, Risks, and Best Practices

Abstract

Artificial intelligence (AI) is increasingly embedded in managerial decision-making, from resource allocation and performance evaluation to strategic planning and customer relationship management. While AI promises efficiency, accuracy, and scale, its integration into management practice raises questions about trust, accountability, job redesign, and ethical risk. This study examines how perceived usefulness of AI, trust in AI systems, managerial AI literacy, perceived risk, and organizational support jointly shape human-AI collaborative performance. Using a cross-sectional survey of 410 practicing managers across five industry sectors, data were analyzed with SPSS 28 for descriptive and reliability statistics and SmartPLS 4 for partial least squares structural equation modeling (PLS-SEM). Results indicate that perceived usefulness (β = 0.312), trust in AI (β = 0.284), and AI literacy (β = 0.198) significantly and positively predict collaborative performance, while perceived risk exerts a significant negative effect (β = -0.176). Organizational support strongly predicts trust in AI (β = 0.401) and moderates the AI literacy–trust relationship. The model explains 47% of variance in collaborative performance. The study contributes an integrated framework linking technology acceptance and risk-perception literatures to human-AI teaming outcomes, and offers evidence-based best practices for managers and organizations seeking to design effective, responsible human-AI collaboration.