From AI Adoption to Human–AI Collaboration: A Conceptual Framework for Managerial Decision-Making and Organizational Performance
Artificial intelligence (AI) is rapidly entering managerial work, yet adopting AI does not automatically improve decision-making. Recent meta-analytic evidence indicates that human–AI combinations do not consistently outperform the better-performing human or AI working alone, particularly in decision-making tasks. Organizational value therefore depends not merely on AI adoption but also on how humans and AI systems collaborate. Research on AI adoption, trust, AI literacy, and human–AI teaming remains fragmented, highlighting the need for an integrative framework linking these factors to managerial and organizational outcomes. Drawing on an integrative review of recent studies on human–AI collaboration, trust in AI, AI literacy, and managerial decision-making, this conceptual paper develops a framework grounded in Sociotechnical Systems Theory and a human–AI complementarity perspective. The framework proposes that AI capability may influence organizational performance indirectly through human–AI collaboration quality, which in turn shapes managerial decision-making quality. AI literacy and calibrated trust are positioned as enabling conditions. AI literacy helps organizational members interpret AI-generated outputs and recognize system limitations, while calibrated trust supports appropriate reliance on AI recommendations. The framework further proposes that collaboration quality is strongest when trust aligns with AI reliability, whereas overreliance and under-reliance may undermine decision quality. Task uncertainty and governance mechanisms for human oversight and accountability are identified as potential boundary conditions. Note: This is the full-length working paper extending the conceptual summary presented at the International Symposium on New Paradigms in Social Sciences and Humanities: Artificial Intelligence and Human (Ankara, Türkiye, 28–29 October 2026).
Authors
- Audi Al-Qaralleh
Publication Details
- Journal
- Zenodo (CERN European Organization for Nuclear Research)
- Published
- 2026-11-14
- DOI
- https://doi.org/10.5281/zenodo.24231713
- Primary Topic
- AI and HR Technologies
- Type
- preprint