Rethinking managerial rationality in the age of AI: a human–machine collaboration perspective on organizational decision-making

Purpose This paper reconceptualizes managerial rationality in artificial intelligence (AI)-augmented decision-making through the notion of algorithmic-bounded rationality (ABR). It argues that AI does not remove boundedness but relocates it into algorithmic constraints related to data volatility, model opacity and governance maturity. Design/methodology/approach Building on bounded rationality and socio-technical systems theory, this conceptual study develops an ABR framework linking three decision modes (AI-led, human-first and collaborative) to mechanisms of algorithmic boundedness. The framework is further extended through propositions on mode–task fit and governance conditions for sustaining hybrid decision architectures. Findings The analysis shows that human–AI collaboration represents a distinct rationality configuration rather than a midpoint between automation and human judgment. Under ABR, each decision mode becomes effective under different combinations of data intensity, contextual ambiguity and accountability demands. Practical implications Managers should treat AI integration as a redesign of decision governance rather than a technological upgrade, emphasizing appropriate authority allocation and oversight mechanisms. Originality/value The study reframes rationality in the AI era by showing how boundedness shifts from human cognition to socio-technical decision infrastructures. It contributes a mechanism-based framework linking decision modes, task conditions and governance arrangements in AI-augmented decision systems.