Artificial Intelligence Relational Diffusion and Acceptance Theory (AIRDAT): A Comprehensive Framework for Understanding Human-AI Interactions

Abstract

The rapid proliferation of artificial intelligence (AI) technologies has intensified the need for new theoretical frameworks to understand their acceptance and diffusion. Existing technology acceptance models inadequately reflect AI's unique characteristics, necessitating a comprehensive theory. This study develops the Artificial Intelligence Relational Diffusion and Acceptance Theory (AIRDAT) to address this gap. AIRDAT, constructed through systematic literature review and theoretical model development, integrates strengths of existing acceptance models (TAM, UTAUT, DOI, SDT) in the AI context. It comprises five components: Individual Acceptance Factors, Relational and Social Factors, AI-Specific Factors, Contextual and Environmental Factors, and Temporal and Evolutionary Dimensions. The theory's applicability was examined through an education sector case study. Analysis reveals AIRDAT's potential to explain AI technology acceptance and diffusion from a multidimensional, dynamic perspective, emphasizing ethical alignment, human-AI collaboration, and AI literacy. AIRDAT contributes to both theory and practice, offering guidance for AI development, policy-making, and management. The study demonstrates AIRDAT's comprehensive framework for understanding AI acceptance and diffusion, potentially guiding future research, informing policy decisions, and supporting ethical AI system development. However, its complex structure may challenge empirical testing and practical application. Future research should empirically test AIRDAT across cultural contexts and AI types, with longitudinal studies examining temporal aspects of AI acceptance. Research on AIRDAT's implications for AI policy and ethics could contribute to responsible AI governance.