A ZERO-TRUST AND AI-ASSISTED THREAT INTELLIGENCE MODEL FOR CYBER RESILIENCE ACROSS CONVERGED IT-OT ENVIRONMENTS

Abstract

IT-OT convergence improves industrial visibility but creates pathways through which cyber compromise can affect physical operations. This study develops an integrated model linking Zero-Trust architecture, AI-assisted threat intelligence, governance, workforce readiness, regulatory alignment, convergence exposure and cyber resilience. An international pilot dataset of 438 IT, OT and cybersecurity professionals was generated across Germany, the wider European Union, the United Kingdom, North America, Asia-Pacific and the Middle East/Africa. Thirty-one Likert items measured seven constructs. Analysis included reliability, exploratory factor analysis, discriminant validity, robust hierarchical regression, path modelling, bootstrap mediation and moderation, cross-validation and international comparison. Reliability was strong with alpha values from .833 to .894. KMO was .918. Seven factors explained 69.7% of item variance. The resilience model explained 53.2% of variance. Zero-Trust maturity had the largest direct effect (beta = .324) followed by AI-assisted threat intelligence (beta = .222). The indirect Zero-Trust effect through AI capability was .100 with a 95% bootstrap interval from .063 to .142. Governance strengthened the Zero-Trust effect. Germany showed higher regulatory alignment and resilience than the pooled sample. North America recorded the highest average Zero-Trust and AI capability. Regional effects were small and the multivariate test was marginal. The results support a governed architecture in which identity, segmentation, AI intelligence, human validation and regulatory evidence operate together. Empirical validation should target German critical infrastructure and industrial sectors.