Answers are grounded in this paper and cite their sources.
This study examines the nonlinear association between digital management readiness and functional artificial intelligence (AI) integration, with particular attention to changes in marginal associations across the readiness continuum and heterogeneity among enterprise-size classes. Digital readiness is operationalized through the multidimensional Digital Management Readiness Index (DMRI), encompassing digital infrastructure, digital management, human capital, and data and security capabilities. Functional AI Integration Index (FAII) captures the extent of AI use across enterprise functions. The empirical analysis applies generalized additive models and marginal-effects estimation to identify nonlinearities without imposing predefined readiness thresholds. The results demonstrate a nonlinear relationship between DMRI and FAII. The nonlinear specification explains 88.1% of the variation in functional AI integration, compared with 78.7% under the linear model. The marginal association increases from 7.477 at the 5th percentile of DMRI to 72.079 at the 95th percentile, indicating an approximately 9.6-fold increase across the observed readiness continuum. No evidence of saturation is identified within the examined range. Enterprise-size analysis further indicates that differences among small, medium-sized, and large enterprises are largely associated with their unequal positioning along the readiness continuum. Size-specific nonlinear functions provide no material improvement over a common readiness trajectory. The findings establish digital readiness as a continuous capability configuration whose relevance for AI integration intensifies as digital capabilities accumulate. The study contributes a nonlinear, capability-position perspective to the literature on enterprise AI integration and suggests that readiness-based segmentation can complement conventional firm-size criteria in digitalization and AI-support policies.
No takes yet. Share an insight, caveat, or question.
Lyulyov et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: