Artificial Intelligence and Digital Innovation in Smart Tourism Destination Management: An Integrative Review and the AI-SDM Framework

Background: Artificial intelligence (AI) is transforming tourism destinations through personalized services, conversational and generative interfaces, predictive analytics, and data-driven decision-making. Yet technology adoption alone does not explain how destinations convert AI-generated information into coordinated managerial action. Objective: This study addresses that organizational problem through a conceptual integrative review of research on AI-enabled tourism services, smart destination management, destination governance, visitor experience, organizational capabilities, workforce readiness, and responsible data practices. Methods: The synthesis identifies four interconnected application domains: AI-driven personalization, conversational and generative visitor interaction, predictive visitor-flow analytics, and data-driven destination governance. These domains are abstracted into the AI-Enabled Smart Destination Management (AI-SDM) framework, which connects AI-Enabled Service Capability, Smart Destination Governance Capability, and Visitor Experience Outcomes. Results: Visitor and operational intelligence functions as the principal conversion input, while visitor-generated behavioral and experiential feedback supports subsequent organizational learning and AI adaptation. Workforce AI Readiness and Responsible Data Governance are positioned as enabling conditions. The framework contributes to destination-management theory by shifting attention from AI adoption toward the organizational capability required to sense relevant visitor information, interpret and coordinate responses, and responsibly convert intelligence into destination action and visitor value. It complements broad smart-tourism and emerging cognitive-destination architectures by specifying the organizational intelligence-to-action mechanism that links AI-enabled interactions with destination governance and learning. Conclusion: The paper concludes with propositions and an empirical research agenda for measurement development and validation.