Generative Artificial Intelligence in Business Decision-Making: Emerging Frameworks, Applications, and Future Challenges

Abstract

Enterprise adoption of generative artificial intelligence (GenAI) has outpaced the development of theoretical frameworks capable of explaining why some organizations successfully integrate the technology into decision-making while others fail to realize comparable value, leaving much of the applied literature to borrow, adapt, or extend theoretical apparatus developed for earlier waves of information technology adoption. This paper reviews the theoretical frameworks and applied evidence base for GenAI in enterprise decision-making, organizing the literature around five analytical lenses: individual-level technology acceptance theory, organizational-level technology-organization-environment adoption theory, the dynamic capabilities framework for strategic integration, the prediction-machine economic framework, and the recently formalized "jagged technological frontier" capability-boundary framework. The review synthesizes foundational adoption theory with large-scale field experimental evidence on GenAI's productivity effects, and with the governance and ethical-framework literature addressing accountability and explicability in AI-assisted enterprise decisions. Distinct comparative tables map each framework onto its unit of analysis and central explanatory variable, cross-reference specific business functions against the framework best suited to explain observed adoption patterns in each, and organize the field's principal future challenges by the theoretical gap each challenge exposes. The paper concludes that no single framework adequately explains GenAI's enterprise decision-making effects across all analytical levels, and that individual-, organizational-, and task-level frameworks must be combined rather than treated as competing explanations, identifying multi-level theoretical integration as the central future research prospect.