This paper addresses a wicked problem in algorithmic consumer culture, as originally theorized by Rittel and Webber and applied to macromarketing by Wooliscroft: AI-enhanced personalization systems, designed to serve consumer preferences, frequently undermine the consumer agency that constitutes the foundation of market emancipation. We adopt Laczniak and Murphy's conception of socially responsible marketing as marketing practice that prioritizes stakeholder welfare, ethical conduct, and societal benefit over narrow firm interests. Drawing upon authenticity theory, Nissenbaum's contextual integrity framework, an updated privacy calculus framework, psychological reactance theory, and posthumanist perspectives on human-AI relations, we develop the Consumer Emancipation Framework (CEF) to explain when AI liberates consumers and when it oppresses them. The framework identifies two competing pathways. The Emancipatory Pathway operates when algorithmic transparency is high, relationship authenticity is preserved, and privacy-personalization tension remains low; under these conditions, AI amplifies consumer agency by reducing information asymmetry and enhancing autonomous choice. The Oppressive Pathway activates when transparency is absent, authenticity is violated, and privacy costs exceed personalization benefits; here, AI degrades consumer agency through manipulation, surveillance, and dark patterns that exploit cognitive vulnerabilities. We introduce three constructs essential for understanding consumer welfare under algorithmic governance: algorithmic transparency, relationship authenticity, and privacy-personalization tension. Seven propositions specify boundary conditions, with particular attention to how algorithmic literacy, consumer vulnerability, and structural inequalities moderate these relationships. Following Sharma et al. multi-level framework for socially responsible consumption, we map these constructs onto consumer-centric, firm-centric, product-centric,