AI is shifting governance from welfare-era and voluntary CSR models toward integrated, AI-enabled institutional governance that embeds monitoring, prediction and coordination across public and corporate domains. Theoretical synthesis contends this evolution reconceives the welfare state and corporate responsibility as components of a unified, sustainability-oriented socio-technical architecture.

Main Finding

AI is not merely a tool but a constitutive element of an emerging socio-technical governance architecture: an "AI-driven welfare state" in which public welfare institutions, corporate governance (ESG + algorithmic systems), and sustainable development goals become integrated through AI-enabled continuous monitoring, predictive regulation, and cross-domain coordination. This process represents a multi-stage historical transformation of governance and corporate forms.

Key Points

• Historical stages identified:
  • Industrial capitalism (market-dominant governance)
  • Keynesian welfare capitalism (state-led redistribution and stabilization)
  • Sustainable development-oriented governance (institutionalized ESG and multi-stakeholder norms)
  • Emerging AI-driven welfare state (AI as infrastructural governance)
• Corporate governance evolution:
  • Voluntary CSR → Institutionalized ESG frameworks → AI-enabled algorithmic governance and operationalization of stakeholder mandates
• Role of AI:
  • Functions as structural infrastructure for governance (monitoring, prediction, automated coordination, regulatory feedback loops)
  • Lowers information frictions and enables fine-grained, continuous policy and corporate responses
  • Integrates multiple institutional domains (state, firms, civil society) into unified socio-technical systems
• Theoretical synthesis:
  • Combines Marxist political economy (power, capital accumulation), Keynesian welfare theory (state role in redistribution and stabilization), sustainable development approaches (multi-objective institutionalization), stakeholder governance theory, and AI governance literature to build a multidimensional explanatory model
• Normative/open issues highlighted:
  • Concentration of power and rents around AI infrastructure providers
  • Redistribution and fiscal implications for welfare financing
  • Institutional lock-in, path dependence, and accountability of algorithmic governance

Data & Methods

• Research design: theory-building (conceptual) study
• Methods used:
  • Historical-comparative analysis: tracing governance and corporate forms across historical stages
  • Institutional analysis: examining formal/informal rules, organizational forms, and governance modalities
  • Conceptual synthesis: integrating multiple theoretical traditions into a coherent model
• Empirical basis: qualitative and historical evidence, literature synthesis across political economy, welfare state theory, sustainability governance, and AI governance scholarship
• Limitations (implicit in design):
  • No primary quantitative or causal identification—results are theoretical and interpretive
  • Proposes mechanisms and trajectories rather than providing estimation of magnitudes or counterfactuals

Implications for AI Economics

• Measurement and modeling
  • Need to treat AI as a form of capital/infrastructure in macro and firm-level models (impacts on productivity, capital share, and rents)
  • Develop metrics for AI-enabled governance externalities (information rents, surveillance value, regulatory effectiveness)
  • Incorporate AI-driven coordination and monitoring into dynamic general equilibrium and public-finance frameworks
• Distribution and welfare
  • AI-driven governance can reduce transaction costs and information asymmetries but may concentrate economic power—requiring new redistributive instruments (taxation of AI rents, social insurance for displaced tasks)
  • Re-evaluate labor share, bargaining power, and unemployment risk under algorithmic coordination and platform-mediated labor
• Corporate behavior and market structure
  • AI enables real-time implementation of ESG/ stakeholder objectives—affecting firm investment, disclosure incentives, and competitive dynamics
  • Potential for winner-take-most markets around algorithmic governance platforms—antitrust and industrial policy implications
• Regulatory design
  • Predictive regulation and continuous monitoring create opportunities for more precise policies but raise concerns about accountability, bias, and surveillance
  • Design of algorithmic oversight institutions, data governance regimes, and split incentives between public and private governance will shape economic outcomes
• Research agenda
  • Empirical quantification of AI's contribution to rents, productivity, and fiscal capacity
  • Causal studies on AI-enabled monitoring and welfare outcomes (natural experiments, RCTs where feasible)
  • Macro models that embed algorithmic governance and institutional feedbacks (agent-based models, DSGE with endogenous institutions)
  • Policy experiments on taxation, universal programs, and public provisioning of AI infrastructure to assess distributional effects

Overall, the study argues economists should expand analyses of AI beyond productivity and automation effects to include AI's role as governance infrastructure that reshapes institutions, distributions, and coordination across economies.