Can artificial intelligence enable environmental sustainability? Evidence from a resource-dependent economy

Title : Can artificial intelligence enable environmental sustainability? Evidence from a resource-dependent economy

Abstract:

Artificial intelligence (AI) is increasingly positioned as a transformative tool for addressing environmental challenges through energy optimization, environmental monitoring, renewable-energy integration, and data-driven decision-making. Yet empirical evidence demonstrating whether growing AI capacity translates into measurable environmental improvements at the national level remains limited, particularly in resource-dependent economies. This study investigates the relationship between AI capacity, economic growth, and environmental sustainability using Qatar as an analytical case over the period 2002–2023. A composite AI capacity index is developed to capture the structural conditions enabling AI adoption through three dimensions: digital infrastructure, institutional capacity, and economic capability.

A progressive econometric framework is employed to distinguish direct, controlled, nonlinear, and indirect relationships between AI capacity and environmental performance. The baseline model initially identifies a positive and statistically significant association between AI capacity and CO₂ emissions (β = 0.168, p < 0.001). However, after controlling for economic structure, the AI effect becomes statistically insignificant (β = −0.069, p = 0.367), while GDP emerges as the dominant determinant of emissions. A nonlinear specification provides evidence of a developing Environmental Kuznets Curve, indicating that the relationship between economic growth and environmental pressure begins to flatten at high income levels.

Importantly, AI capacity exhibits a positive, although modest, association with renewable-energy consumption (β = 0.019, p ≈ 0.078), revealing a potential indirect pathway through which AI can support environmental sustainability. The findings therefore challenge technology-centric assumptions that AI adoption automatically generates environmental benefits. Instead, AI emerges as an enabling infrastructure whose environmental contribution depends on its integration with renewable energy, low-carbon infrastructure, economic diversification, and effective environmental governance. The study provides evidence for policymakers seeking to align digital transformation with environmental and energy-transition strategies in resource-dependent economies.

Keywords: Artificial Intelligence; Environmental Sustainability; CO₂ Emissions; Renewable Energy; Energy Transition; Environmental Kuznets Curve; Qatar

Biography:

Faisal Melhim is an AI Engineering student at Ulster University in Qatar with research interests at the intersection of artificial intelligence, data science, environmental sustainability, and sustainable energy. His research includes work on AI capacity and environmental sustainability transitions, digital readiness and energy transition in GCC countries, and IoT-based environmental monitoring. He has presented his research at international conferences and gained research experience at the Qatar Computing Research Institute. Faisal has also contributed to his university’s first Carbon Inventory Project, supporting the assessment of institutional carbon emissions and sustainability initiatives.