Digital platforms now act as institutional gatekeepers, using data and algorithms to concentrate economic rents and shape public consent, even as platform affordances allow limited space for alternative voices.

Main Finding

Platform capitalism functions as a multi-dimensional institutional form that concentrates economic and ideological power. Digital platforms are not mere communication tools but new institutional actors that construct and reproduce power through control of data, algorithms, and digital infrastructure. While this control tends to produce digital hegemony (consent and social order favoring platform interests), social media and platform-embedded practices also create openings for alternative voices and counter-hegemonic movements. A neo‑Gramscian synthesis of economic, political, and cultural analysis is required to fully understand contemporary digital power relations.

Key Points

• Platforms-as-institutions: Platforms have evolved into institutional actors that coordinate markets, norms, and social interactions, shaping possibilities for political and economic action.
• Data and algorithmic control: Ownership and governance of data and algorithms are primary mechanisms through which platforms exert power—shaping information flows, attention allocation, and behavior.
• Algorithmic governance: Algorithms function as governance tools (shaping visibility, social sorting, enforcement of norms) and thus contribute to the reproduction of hegemonic arrangements.
• Ideological hegemony and consent: Platform practices and content dynamics help produce cultural consensus that normalizes platform-centric economic arrangements and power asymmetries.
• Counter-hegemony and contestation: Social media and platform affordances also enable grassroots mobilization, alternative publics, and contestation of dominant narratives—though such spaces are constrained by platform design and political economies.
• Integrated theoretical lens: A neo‑Gramscian approach (combining political economy, ideology/hegemony, and civil society) is proposed as the appropriate framework to analyze these multi-layered power relations.
• Limits acknowledged: The study is theoretical and interpretive, drawing on secondary literature; it synthesizes existing findings rather than generating new quantitative evidence.

Data & Methods

• Methodological approach: Qualitative, theoretical synthesis and critical review.
• Sources used: Secondary literature only—books, journal articles, theoretical texts, and in-depth critical reviews of existing empirical and conceptual research.
• Analytical framework: Neo‑Gramscian political economy used to interpret relationships among economic structures (platform capitalism), ideological production (media, narratives), and civil society (publics, movements).
• Limitations of approach:
  • No primary empirical data or quantitative analysis.
  • Findings rely on interpretation of existing studies and may reflect selection or publication biases in the secondary literature.
  • Causal claims are not empirically tested; generalizability depends on the robustness of cited work.

Implications for AI Economics

• Market power and rents: Control over massive, high-quality data and proprietary algorithms creates strong scale and scope economies for platform firms, enabling persistent market power and rent extraction. Economic models of competition and antitrust must incorporate data-based barriers to entry and algorithmic complementarities.
• Inputs to AI: Data governance determines the availability and quality of training data for AI systems—affecting innovation trajectories, concentration of capabilities, and distribution of economic surplus from AI-enabled services.
• Platform gatekeeping & multi-sided markets: Algorithms mediate access to consumers and workers, altering pricing, matching, and surplus division across sides of platform markets. Platforms can shape match quality and extract commissions/rents through algorithmic control.
• Attention and informational externalities: Algorithmic curation organizes attention and shapes demand, creating externalities (herding, polarization, misinformation) that have market and welfare consequences not captured in standard competitive models.
• Labor economics: Algorithmic management governs gig and platform labor, affecting bargaining power, wage setting, monitoring intensity, and the nature of employment relationships—requiring updated models of labor supply, incentives, and regulation.
• Innovation and path dependence: Platform architectures and entrenched datasets can produce lock‑in effects that bias innovation toward incumbent interests and reduce diversity of technological trajectories.
• Policy and regulation: Findings support the need for interventions (data portability, interoperability, algorithmic transparency, antitrust enforcement tailored to data/algorithmic assets, platform governance rules) that address structural and ideological dimensions of platform power.
• Measurement & empirical research gaps: AI economics should develop metrics for platform hegemony (data concentration indices, algorithmic gatekeeping measures, attention market shares), causal methods to assess algorithmic effects on markets and political outcomes, and microdata access frameworks for independent research.
• Political economy of AI: Economic analysis of AI must incorporate ideological and social dimensions (how AI-mediated narratives shape preferences and consent), implying interdisciplinary methods (economics + political sociology) for policy design and welfare assessment.