Regulating the Environmental Impacts of Artificial Intelligence in the Public Sector

Objective

The project “Regulating the Environmental Impacts of Artificial Intelligence in the Public Sector” (SUSTAIN-IA-PUB) addresses the urgent need to govern the environmental costs of AI, which remain largely invisible and unregulated despite AI’s role as a driver of the digital and green transitions. Current EU legislation, including the AI Act (Reg. EU 2024/1689), pays only marginal attention to sustainability, leaving a regulatory gap that this research seeks to fill. The main objective is to promote environmentally responsible AI adoption in the public sector through the development of methodologies and regulatory strategies that are adaptable, transparent and operational. The project pursues four goals. First, it will map and compare how public administrations in Europe and beyond currently assess the ecological footprint of AI, identifying gaps and best practices. Second, it will design an interdisciplinary methodology, grounded in lifecycle analysis and enriched by legal and technical expertise, to measure both direct impacts such as energy and water use and indirect effects such as rebound phenomena. Third, it will translate these findings into actionable regulatory instruments, including procurement criteria, reporting obligations and soft law tools that public bodies can realistically adopt. Finally, it will engage society through a citizen guide, a policy brief and an open-access repository to raise awareness and empower informed debate. By integrating law, environmental studies, computer science and governance, SUSTAIN-IA-PUB contributes directly to Horizon Europe’s priorities on the twin transitions. It will strengthen the capacity of Member States and public institutions to adopt AI systems that are lawful, effective and ecologically sustainable, positioning environmental responsibility as a core dimension of trustworthy AI and supporting Europe’s ambition to lead the digital and green transformation.