Epistemic Decolonization in the 21st Century: Deconstructing AI, Big Data, And Global Knowledge Systems

Despite promises of technological neutrality, the digital revolution often reproduces—and sometimes intensifies—historical patterns of inequality and exclusion. This paper investigates how artificial intelligence (AI), big data, and global knowledge infrastructures continue to reflect and reinforce colonial hierarchies, frequently sidelining non-Western epistemologies and communities. Grounded in decolonial theory, the analysis draws on diverse case studies, including the MASAKHANE project for African language AI, Indigenous-led data sovereignty initiatives, and policy debates in the Global South, to illustrate both the persistence of digital coloniality and the rise of innovative local resistance. By mapping how data extraction, algorithmic bias, and the epistemic marginalization of minority cultures are embedded in contemporary digital systems, the research reveals the limitations of technical "fixes" and argues for a more fundamental transformation of digital governance. The paper advances a framework for epistemic decolonization that foregrounds pluralism, community control, and the recognition of multiple knowledge systems in technology design and policy. Policy recommendations are offered to support collective data governance, participatory AI development, and capacity building in underrepresented regions. Ultimately, the study calls for a reimagining of digital futures in which justice, inclusion, and epistemic diversity are central—not peripheral—to technological progress. This approach has broad implications for scholars, practitioners, and policymakers seeking to build more equitable, plural, and sustainable digital societies.