Artificial Intelligence and Economic Transformation: Implications for Productivity, Employment, Innovation, and Sustainable Economic Growth

Abstract

This study systematically synthesizes the economic literature on artificial intelligence (AI), productivity, employment, and growth available up to 2024. An organizational framework based on a PRISMA approach is employed to structure 50 leading and citable research papers on productivity, innovation, labor demand, skills, distribution, and macroeconomic growth. The synthesis is based on the task-based technological change theory and the general-purpose technology perspective, with a focus on the differences between worker-, firm-, and economy-level evidence. There is growing evidence in the literature that AI complements work to boost the productivity of workers and firms, or that AI can assist in decision-making or product innovation. Aggregate productivity effects are less well agreed on, due to incomplete diffusion and measurement lags in the effect of complementary investments. Labor-market effects are heterogeneous, as displacement, augmentation, and new-task creation happen at the same time, and skill, occupation, sector, and firm characteristics condition exposure. The review integrates microeconomic, labor, and macroeconomic evidence in a single framework and outlines the research agenda on diffusion, developing economies, causal identification, generative AI, inequality, and long-run growth.