The use of artificial intelligence (AI) in the digital economy has drastically changed business processes and public services. However, in another way, this technology further exacerbates social and economic disparities. Disadvantaged groups such as low-skilled workers, micro and small enterprises (MSMEs), and groups less accessible to digital technology are largely not equally endowed with AI technology and digital infrastructure. This research aims to establish a new theoretical framework for comprehending the social and economic impacts of AI uptake in the digital economy, particularly on vulnerable groups. Employing a qualitative case study approach, this study is literature-reviewed and document-analyzed and based on five sociological theories at its essence: Social Stratification, Social Inequality, Social and Cultural Capital, Modern Stratification, and Network Society. The results of the study show that utilization of AI works to benefit individuals or groups with digital literacy skills and technological access, while reinforcing marginalization of those with fewer resources. This situation amplifies inherent structural inequality and creates a new layer in the form of digital stratification. The conceptual framework derived from this study presents an integrated and multi-disciplinary way of comprehending the far-reaching social implications of AI implementation. Apart from identifying the potential setbacks of technological exclusion, this research is also a springboard upon which to design more equitable and inclusive AI policy. By connecting classic sociological theory to present-day digital dynamics, this research presents a new contribution in the guise of analytical tools for assessing justice and inclusion in an AI economy.