AI inequality fuels cross-border pollution and environmental injustice

The authors frame this as the “AI paradox” - a condition where digital progress, if unevenly shared, generates both innovation and injustice. They argue that bridging the AI gap is essential not only for economic competitiveness but also for environmental stability.

A new study published in Sustainability has raised urgent questions about the unintended environmental consequences of artificial intelligence (AI) progress. Conducted by Ran Cui, Pengfei Zhao, Qingfeng Luo, and Jingyuan Wang, the research finds that unequal technological advancement in AI not only drives economic disparity but also worsens air pollution transmission across regional borders.

Titled “The Artificial Intelligence Paradox: Does Digital Progress Fuel Environmental Injustice via Transboundary Pollution?”, the study provides compelling empirical evidence that AI development, when unevenly distributed, can intensify environmental injustice by amplifying cross-city pollution flows. Drawing from nearly 98,000 paired observations across 157 Chinese cities between 2015 and 2022, the authors reveal a troubling paradox: technological innovation, while essential for growth, may simultaneously deepen ecological inequality.

Unequal AI growth and its hidden ecological costs

According to the study, AI is not just as a technological revolution but a complex socio-environmental factor with far-reaching implications. Grounded in Technology Gap Theory (TGT), the research argues that differences in AI capability, known as the AI gap, create disparities in production efficiency, industrial upgrading, and environmental governance. Cities at the forefront of AI development tend to attract more digital infrastructure investment, high-skilled labor, and capital inflows, creating a feedback loop of progress. In contrast, lagging regions experience slower technological diffusion and reduced capacity to enforce environmental standards.

This widening AI gap, according to the authors, significantly increases the intensity of transboundary air pollution, meaning emissions from one city are more likely to affect neighboring areas. AI-advanced cities benefit from cleaner industries and stronger regulation, while less advanced regions become the “pollution receivers,” shouldering the environmental burden of industrial relocation.

The findings underscore how AI can exacerbate environmental asymmetry when digital progress is concentrated in economically powerful areas. As smart technologies enhance production efficiency in developed cities, industries reliant on heavy emissions often relocate to less technologically equipped regions where governance mechanisms are weaker. This industrial transfer effect reinforces a two-speed model of progress, technological in some regions and ecological decline in others.

In essence, while AI has the potential to accelerate green innovation, the study finds that unequal diffusion of AI resources amplifies pollution spillovers, contributing to environmental injustice between high-tech and low-tech cities.

Mechanisms behind the AI gap and pollution transmission

The researchers identify three key mechanisms driving this paradox: digital infrastructure disparities, labor mobility imbalances, and capital flow asymmetries.

First, digital infrastructure plays a crucial role in determining how well cities can manage environmental challenges. AI-leading cities invest heavily in data-driven environmental monitoring, automated emissions tracking, and smart grid systems. These tools enable rapid responses to pollution events. Conversely, AI-lagging regions lack such systems, resulting in slower detection and response to air quality deterioration.

Second, labor mobility contributes to environmental divergence. The study finds that skilled workers are drawn to AI-intensive cities offering higher wages and better living standards. This brain drain deprives less developed areas of technical expertise essential for managing sustainable growth. Consequently, low-tech regions struggle to implement green innovations and continue relying on outdated, pollutive industries.

Third, capital flow exacerbates inequality. Investors prioritize AI-advanced cities with robust technological ecosystems, further entrenching economic concentration. The lack of green investment in underdeveloped areas constrains their ability to transition toward cleaner production models, reinforcing their dependence on high-emission activities.

Collectively, these mechanisms create a cumulative disadvantage, where the AI gap not only limits economic opportunity but also exposes less developed regions to greater environmental risk. The researchers describe this as a self-perpetuating cycle, where technological inequality breeds ecological inequality.

The study further reveals that narrowing the AI gap can mitigate air pollution transmission. When technological capabilities converge, cities develop shared digital governance systems that facilitate real-time environmental coordination. This convergence leads to lower pollutant transfer between urban centers, highlighting that digital inclusivity is directly linked to environmental equity.

Policy responses and the path toward sustainable AI governance

The research provides critical insights for policymakers confronting the environmental side effects of digital transformation. Through heterogeneity analysis, the authors find that the negative environmental impacts of AI inequality are most severe in economically underdeveloped regions, where limited governance capacity and weak enforcement amplify pollution transfer.

Cities participating in low-carbon pilot programs demonstrate significantly reduced pollution transmission, suggesting that targeted environmental governance can offset the damage caused by digital disparities. These programs promote the adoption of cleaner energy systems and integrate digital monitoring tools, effectively bridging the gap between AI and environmental management.

On the other hand, initiatives like the Smart City Pilot Program, while boosting technological growth, may unintentionally worsen environmental inequality if not aligned with green policies. The study argues that digital progress without ecological safeguards risks magnifying existing disparities, as infrastructure and innovation flow predominantly toward already-advantaged cities.

The authors call for a dual-governance model that unites digital and environmental policy frameworks. This approach would ensure that technological advancement contributes to sustainable outcomes rather than ecological harm. Specifically, they advocate for:

• Balanced AI infrastructure investment across regions to prevent concentration of digital power.
• Green-oriented technology diffusion that promotes industrial upgrading in lagging areas.
• Cross-regional coordination mechanisms for managing transboundary air pollution through data-sharing and AI-driven analytics.

The study emphasizes that AI governance must be both inclusive and environmentally conscious, integrating economic, social, and ecological priorities. Addressing digital inequality is not only a technological imperative but a necessary condition for achieving environmental justice.

Global implications: The AI-environment nexus

While the research is based on Chinese data, its implications extend globally. As nations race to deploy AI for productivity and sustainability, the uneven distribution of digital capacity threatens to reproduce the same inequalities seen in industrial development eras. Regions with robust AI ecosystems will gain efficiency and resilience, while others risk environmental degradation as industrial displacement intensifies.

The authors frame this as the “AI paradox” - a condition where digital progress, if unevenly shared, generates both innovation and injustice. They argue that bridging the AI gap is essential not only for economic competitiveness but also for environmental stability.

To sum up, sustainable technological advancement requires inclusive digital strategies that consider both economic growth and ecological equity. By aligning AI innovation with environmental governance, policymakers can transform AI from a driver of disparity into a catalyst for shared sustainability.