AI promises sizable productivity and GDP gains for ready emerging-market economies, but the windfall is conditional; India — led by hubs like Bengaluru — exemplifies the upside, while inadequate infrastructure, skills gaps and governance shortfalls could blunt benefits and exacerbate inequality.

Main Finding

AI can be a powerful accelerator of economic growth in emerging economies, but its benefits are conditional. The paper’s narrative synthesis of 35 studies (2018–2026) reports that AI adoption in high-readiness emerging economies can raise total factor productivity (TFP) by roughly 15–25% and add an estimated 1–3% to annual GDP growth. India — especially Karnataka/Bengaluru and initiatives like India AI Mission and AgriStack — is presented as a leading example of how digital ecosystems and regional innovation clusters can translate AI into inclusive sectoral gains (notably in agriculture, manufacturing and services). However, substantial constraints (digital divide, skills gap, job displacement, governance and ethical risks) mean policy choices determine whether gains are broad-based or reinforce inequality.

Key Points

• Quantitative claims from the review
  • AI adoption → TFP increases ~15–25% (reported range across studies).
  • Potential contribution to annual GDP growth in high-readiness EMEs: ~1–3%.
  • Precision agriculture applications reported yield gains ~15–25% and reduced resource usage.
• Sectoral impacts
  • Agriculture: precision farming, AgriStack → better yields, input efficiency, rural inclusion potential.
  • Manufacturing: automation/robotics → higher efficiency and industrial productivity gains.
  • Services: AI analytics/chatbots → improved customer experience and operational efficiency.
• Geographic and institutional examples
  • India: large digital user base, Digital India, India AI Mission; AgriStack as a use-case.
  • Karnataka/Bengaluru: regional innovation hub linking advanced AI with rural/agricultural applications.
• Risks and distributional concerns
  • Digital divide: rural/low‑connectivity areas lag, limiting adoption.
  • Skills gap and informality: risk of widening inequality as low‑skilled workers face displacement.
  • Ethical/regulatory issues: data privacy, algorithmic bias, accountability need governance.
• Policy prescription themes
  • Invest in digital infrastructure (e.g., BharatNet), human capital (reskilling), governance (data/privacy/ethics), social protection, and public–private partnerships.

Data & Methods

• Review type: PRISMA-ScR–guided narrative review (systematic in selection, narrative in synthesis).
• Evidence base: 35 empirical studies published 2018–2026, supplemented with government reports, international organizations and institutional sources.
• Data sources cited include BIS (2025), McKinsey (2023), NITI Aayog/India AI Mission (2024–2025), World Bank, OECD and state-level (Karnataka) reports.
• Analytical approach: thematic analysis to identify trends; comparative analysis across contexts; descriptive synthesis of sectoral and policy findings.
• Modeling and methodological references: mentions use of computational modelling, clustering (location-based), fuzzy decision reasoning and welfare-perspective analyses in supporting literature [refs 8–11].
• Limitations (noted or implied): reliance on secondary sources and heterogeneous empirical designs; narrative synthesis limits causal inference and may reflect publication/selection biases.

Implications for AI Economics

• Measurement & macro implications
  • Need for refined metrics of AI capital and AI-driven TFP to quantify growth contributions more robustly.
  • Incorporate AI adoption heterogeneity (readiness, infrastructure, skills) into cross-country growth models and simulations.
• Labor market and distribution
  • Model short- and long-term labor reallocation effects: displacement vs. new task creation; incorporate informality and limited mobility in emerging economies.
  • Evaluate welfare and inequality impacts; design counterfactuals for reskilling, social protection and job-creation policies.
• Policy design and evaluation
  • Prioritize investments where returns to AI are inclusive: connectivity, local data infrastructure, targeted upskilling and sectoral pilots (e.g., agriculture).
  • Design regulatory frameworks for data governance, algorithmic accountability and privacy that are compatible with local institutional capacity.
  • Use regional innovation clusters (Karnataka/Bengaluru model) to test scalable policies linking advanced R&D with rural applications.
• Research agenda
  • Empirical: more causal microstudies (randomized or quasi-experimental) on AI interventions’ productivity, employment and welfare effects in EMEs.
  • Longitudinal studies on inequality and dynamic labor market outcomes as AI diffuses.
  • Cross-country work to identify thresholds of “readiness” and policy complementarities that translate AI into broad-based growth.
  • Incorporate computational and welfare-oriented models (as referenced) to assess distributional consequences and optimal policy mixes.
• Institutional & international cooperation
  • Encourage international standards and knowledge transfer for data governance, benchmarking AI readiness, and capacity building in governance and regulation.

Overall, the paper underscores that AI’s macroeconomic promise in emerging economies is substantial but not automatic: economists and policymakers must couple measurement, causal evidence and targeted policy instruments to ensure growth is both productive and inclusive.