Regional AI infrastructure choices inevitably trade off rapid progress, environmental sustainability, and equitable outcomes; absent deliberate governance, market power and regulatory inertia will default allocation toward narrow priorities.

Main Finding

The paper introduces the "AI Infrastructure Triad"—Progress, Sustainability, and Equity—as a practical framework for regional AI infrastructure governance, and argues that under current technological, institutional, and resource constraints regions are unlikely to maximize all three goals simultaneously. It shows how stakeholder comments and regional positions reflect different prioritized frontiers of the triad, clarifies the trade-offs infrastructure decisions obscure, and proposes a Deliberate Triad Choice Framework to guide policymakers away from default, market- or inertia-driven allocations.

Key Points

• AI Infrastructure Triad
  • Progress: rapid deployment of compute, data centers, and associated services to accelerate AI development and capture economic gains.
  • Sustainability: minimizing environmental footprints (energy, carbon, water, land) and long-term resilience of supporting systems.
  • Equity: fair distribution of costs and benefits across places, workers, and communities, including environmental justice and labor impacts.
• Trade-offs
  • Investments that maximize Progress (e.g., large-scale, cheap compute) often increase energy, water, and land use, conflicting with Sustainability or placing burdens on local communities.
  • Prioritizing Sustainability (e.g., strict environmental limits or siting constraints) can slow deployment and raise costs, reducing Progress and potentially limiting economic opportunity.
  • Prioritizing Equity (e.g., community control, local hiring, stringent safeguards) can alter siting and operational choices, raising trade-offs with both Progress and cost-efficiency.
• Limits and constraints
  • The paper draws on prior work identifying economic, physical, and moral limits to AI expansion (e.g., grid capacity, water scarcity, labor market impacts, ethical/justice considerations).
  • These limits help explain why simultaneous maximization of all triad goals is unlikely under present conditions.
• Governance contrast
  • Distinguishes deliberate triad governance—explicitly choosing which priorities to emphasize—from default allocation driven by market power or regulatory inertia, which tends to favor Progress unless countered.
• Practical policy tool
  • Proposes a Deliberate Triad Choice Framework to help regional policymakers make explicit, transparent choices among competing priorities when evaluating large-scale AI infrastructure decisions.

Data & Methods

• Evidence sources
  • Synthesis of prior literature on the economic, physical, and moral limits of AI development.
  • A previously coded dataset of 10,068 public comments submitted to the 2025 U.S. AI Action Plan to characterize stakeholder positions and how they map to triad priorities.
  • Illustrative regional case examples to show how triad trade-offs manifest in different local contexts.
• Analytical approach
  • Conceptual framing: develops the triad and articulates frontier trade-offs rather than estimating a single causal model.
  • Interpretive mapping: uses coded public comments and regional illustrations to interpret stakeholder and jurisdictional priorities as choices among triad frontiers.
  • Evidence is used illustratively to clarify patterns and governance options; the paper does not claim full causal identification or exhaustive empirical testing.

Implications for AI Economics

• Reframe economic analysis
  • Move beyond a simple "innovation vs regulation" binary to multi-objective analyses that explicitly include Sustainability and Equity when valuing AI infrastructure investments.
  • Incorporate resource constraints (grid capacity, water, land) and distributional outcomes into cost-benefit and welfare models of AI rollout.
• Policy design and evaluation
  • Encourage policymakers to adopt Deliberate Triad Choices—explicit, transparent trade-off decisions—rather than relying on market outcomes or regulatory inertia.
  • Tailor instruments to chosen priorities: e.g., subsidies or fast permits for Progress; stricter siting, emissions, or water-use limits for Sustainability; community benefit agreements and labor protections for Equity.
• Research directions for economists
  • Quantify trade-offs empirically: spatial equilibrium models, social-cost accounting for environmental externalities, distributional impact assessments.
  • Develop metrics that operationalize Progress, Sustainability, and Equity for comparative policy evaluation.
  • Study market power and regulatory capture dynamics that produce default triad allocations, and test the causal effects of deliberate policies on infrastructure outcomes.
• Practical implications for regions
  • Regions should assess local resource endowments and constraints, stakeholder priorities, and institutional capacity to pursue explicit triad choices aligned with local objectives.
  • Anticipate coordination needs across energy, water, land-use, labor, and environmental agencies to manage cross-cutting impacts of large AI infrastructure projects.