The expansion of artificial intelligence is raising an alert regarding water consumption in Brazil. A UNU-INWEH study reveals that the Brazilian electricity matrix, heavily reliant on hydropower, requires increased attention to environmental and energy management for data centers.
The expansion of artificial intelligence (AI) brings not only technological and economic advancements but also deepens challenges related to the consumption of natural resources. In Brazil, this scenario takes on a particular dimension due to the characteristics of its electricity matrix. A recent report by the United Nations University Institute for Water, Environment and Health (UNU-INWEH) indicates that the Brazilian electricity matrix exhibits a water use intensity 191% higher than the global average.
This data is crucial when considering the exponential growth of infrastructure needed to support artificial intelligence systems, especially data centers. Globally, these structures demanded approximately 448 TWh of electricity in 2025, a volume that could jump to 945 TWh by 2030. The annual water demand associated with this sector also projects concerning figures, potentially reaching 9.3 trillion liters.
Although Brazil prides itself on a predominantly renewable electricity matrix, the strong reliance on hydroelectric generation places water use as a significant environmental factor in evaluating the sustainability of AI expansion. The UNU-INWEH study emphasizes that analyzing the sustainability of artificial intelligence solely from the perspective of carbon emissions can be an incomplete approach, disregarding other equally relevant environmental impacts.
Challenges and Solutions for AI Sustainability
Given this scenario, the Brazilian Association for Quality Infrastructure (ABRIQ) underscores the importance of establishing clear technical and regulatory criteria. These guidelines are fundamental for the growth of the artificial intelligence sector to occur in an environmentally responsible manner, ensuring that technological progress does not compromise natural resources.
For Alexandre Xavier, ABRIQ’s Vice President of ESG, technological development must be intrinsically accompanied by effective mechanisms for measuring and reducing impacts. He reinforces the need for a more comprehensive vision of sustainability, going beyond energy consumption and carbon emissions.
“The expansion of artificial intelligence needs to be accompanied by a broader vision of sustainability. In addition to energy consumption and carbon emissions, it is fundamental to establish parameters that allow for measuring water usage, evaluating system efficiency, and monitoring their impacts on natural resources,” states Xavier.
International Standards: Guides for Environmental and Energy Management
In this context, the adoption of international standards, such as ISO 14001 (Environmental Management System) and ISO 50001 (Energy Management System), presents itself as a promising path. These frameworks can structure organizations’ environmental and energy management, enabling the establishment of robust monitoring processes, the definition of clear goals, and the continuous identification of opportunities for improving resource use efficiency.
Alexandre Xavier highlights the benefits of these tools for technology companies and data center operators. The implementation of environmental and energy management systems can assist in creating performance indicators, allowing for precise monitoring and the identification of concrete actions to reduce environmental impacts in a rapidly accelerating sector.
“The adoption of environmental and energy management systems can help technology companies and data center operators establish indicators, monitor their performance, and identify concrete opportunities to reduce environmental impacts. This is especially important in a rapidly growing sector that demands increasingly efficient management of its resources,” notes Xavier.
Quality Infrastructure: A Pillar for Conscious Growth
ABRIQ emphasizes that Quality Infrastructure plays a strategic role in this scenario. Through tools such as standardization, conformity assessment, certification, accreditation, and metrology, it is possible to contribute to the creation of reliable criteria for environmental performance and efficiency. These tools are essential to ensure that AI expansion in Brazil is carried out in a responsible and measurable manner.
Brazil has a unique opportunity to leverage the advancement of artificial intelligence without compromising its valuable natural resources. For this to materialize, the creation of clear rules, the development of reliable indicators, and the implementation of effective verification mechanisms are indispensable. These elements are crucial for monitoring the sector’s impacts and for stimulating the emergence of more efficient and sustainable solutions.
“Brazil has the opportunity to leverage the growth of artificial intelligence without sacrificing the protection of its natural resources. To do so, we need clear rules, reliable indicators, and verification mechanisms that allow for monitoring the sector’s impacts and stimulating more efficient and sustainable solutions,” concludes Alexandre Xavier.
