The Structural Contradictions of Capitalist AI

This study critically examines systemic contradictions in artificial intelligence (AI) under capitalism through a Marxist framework, with epistemic enclosure – the private monopolization of knowledge, data, and computational infrastructure – as its central concept. It identifies three interlocking contradictions: technological monopolization, digital labor exploitation, and epistemic colonialism. Drawing on Marx's theories of primitive accumulation and surplus value, the analysis shows how corporate control over data, infrastructure, and knowledge production entrenches inequality and alienation. Empirical cases – ranging from digital enclosures and labor exploitation in the Global South to Western epistemic hegemony – illustrate the symbiotic dynamics of these contradictions. The study contends that ethical reforms cannot resolve these systemic flaws and calls for a paradigm shift toward AI Knowledge Socialism: a framework promoting collective ownership of digital commons, democratic governance of technology, and pluralistic knowledge production to decolonize AI. This vision redefines AI as a collectively governed resource serving social needs while acknowledging unresolved gaps, including environmental costs, gendered labor exploitation, and the democratization of foundational infrastructures. Addressing these areas is essential for a comprehensive critique of capitalist AI.