Modeling AI-driven inequality and adaptive governance: A system dynamics approach to U.S. Socioeconomic futures

Artificial Intelligence is reshaping socioeconomic systems by enhancing productivity while intensifying concerns about inequality, unemployment, and policy responsiveness. This study employs a System Dynamics model to simulate the U.S. socioeconomic landscape from 2000 to 2035, focusing on the interdependencies between AI investment, income distribution, and adaptive policy design. Given data constraints, AI investment is modeled as a uniform labor market driver, with international competition introduced via the DeepSeek stress test. The model integrates political feedback loops linking wealth concentration to reform inertia. Three policy scenarios are evaluated: (1) baseline U.S. AI adoption, (2) competitive pressure from a low-cost foreign platform under varying regulations, and (3) adaptive reforms coupling AI taxation and redistribution to real-time inequality and unemployment metrics. Results reveal that while AI-driven productivity may reduce unemployment and cost of production initially, it exacerbates inequality without responsive governance. Adaptive mechanisms, such as dynamic reskilling and AI-linked fiscal tools, outperform static interventions in promoting equity and competitiveness. However, entrenched political influence constrains reform unless public dissatisfaction crosses critical thresholds. These findings highlight the urgent need for anticipatory, adaptive policy frameworks that align technological innovation with inclusive and sustainable socioeconomic outcomes.