Prioritizing barriers to AI adoption in healthcare enterprise systems: a mixed-methods approach to governance and readiness

Abstract

Purpose Despite immense growth in artificial intelligence (AI) within healthcare enterprise systems, early-stage adoption remains furrowed. By privileging technical capability and individual acceptance, existing research underexplains AI adoption in professionalized, high-stakes contexts where accountability structures, autonomy and decision rights shape outcomes. This study examines early AI adoption through a governance-centric lens by framing it as an enterprise information governance challenge. Design/methodology/approach A mixed-methods design was adopted to interview physicians working in private hospitals across three Arab Peninsula countries undergoing rapid digital transformation. The grounded theory method was used to identify physician-perceived organizational, technological and people barriers to AI adoption, followed by the Bradley-Terry Model to rank their relative salience using expert pairwise comparisons. Findings Governance-related barriers were more salient than technological or people-related barriers. These barriers function as mechanisms that structure interactions among enterprise technologies, professional expertise and accountability regimes. Organizational readiness emerged configurationally, which reflected alignment among decision rights, accountability structures and professional roles rather than additive readiness factors. Originality/value This study advances enterprise information management theory by positioning AI governance as a central dimension of organizational readiness in professionalized settings. It refines the technology-organization-environment framework and institutional perspectives by foregrounding governance mechanisms. It also offers practical guidance for designing governance arrangements that enable AI integration while preserving professional autonomy and organizational control.