Rising Emissions, Depleting Water and Vanishing Land—UN Scientists Warn AI Is Threatening Natural Resources for Billions
The Environmental Cost of Artificial Intelligence Comes Into Focus
Artificial intelligence is often celebrated as the technology that will define the 21st century, revolutionizing healthcare, education, finance, manufacturing, transportation, and scientific research. Yet a growing body of scientific evidence suggests that the AI revolution carries significant environmental costs that are often overlooked.
According to researchers associated with the United Nations University, the rapid expansion of AI infrastructure is contributing to rising carbon emissions, increasing freshwater consumption, and growing pressure on land resources. Scientists warn that without stronger sustainability measures, the environmental footprint of AI could affect billions of people who depend on already stressed natural resources.
AI’s Hidden Environmental Footprint
Most users experience artificial intelligence through chatbots, recommendation systems, image generators, and digital assistants. Behind these services lies a vast physical infrastructure consisting of data centres, power plants, communication networks, semiconductor factories, and industrial supply chains.
Every AI query requires computing power. Every AI model requires training. Every digital interaction relies on servers operating continuously across thousands of facilities worldwide.
As AI adoption accelerates, the demand for infrastructure continues to expand at an unprecedented pace.
Rising Carbon Emissions
One of the most significant concerns involves energy consumption.
Training advanced AI models requires enormous computational resources, often involving thousands of specialized processors operating for weeks or months. Once deployed, AI systems continue consuming energy as millions of users interact with them daily.
Researchers warn that growing electricity demand from AI data centres could substantially increase greenhouse-gas emissions, particularly in regions where power generation remains dependent on fossil fuels.
Although technology companies are investing heavily in renewable energy, experts caution that AI’s growth may outpace clean-energy deployment in some markets.
Why It Matters
Higher emissions contribute to:
- Global warming
- More frequent heatwaves
- Extreme weather events
- Droughts
- Flooding
- Agricultural disruption
- Public-health challenges
As a result, AI’s carbon footprint has become a major focus of environmental research.
Depleting Freshwater Resources
Water is emerging as one of the most controversial aspects of AI development.
Data centres generate enormous amounts of heat and require cooling systems to maintain safe operating temperatures. Many cooling technologies depend on water, either directly through evaporation-based systems or indirectly through electricity generation.
Researchers estimate that global AI-related water consumption could reach billions of cubic metres annually as adoption expands.
Water Stress Is Already Growing
The challenge is particularly serious because many AI data centres are being built in regions facing:
- Groundwater depletion
- Recurring droughts
- Population growth
- Agricultural water shortages
- Climate-related water stress
Countries such as the United States, India, China, and several Middle Eastern nations are simultaneously expanding AI infrastructure while confronting growing water-security concerns.
For communities dependent on wells, rivers, and aquifers, increasing industrial water demand raises difficult questions about long-term sustainability.
Vanishing Land Resources
Another often-overlooked consequence of AI growth involves land use.
The AI ecosystem requires:
- Data centres
- Power-generation facilities
- Transmission infrastructure
- Semiconductor manufacturing plants
- Warehouses
- Industrial parks
- Mining operations for critical minerals
Together, these developments consume large areas of land and can alter ecosystems.
Critical Minerals and Mining
AI hardware depends on minerals such as:
- Lithium
- Cobalt
- Nickel
- Copper
- Rare earth elements
Extracting these resources often requires extensive mining operations that can affect forests, biodiversity, water systems, and local communities.
Scientists warn that increasing demand for AI hardware may intensify pressure on environmentally sensitive regions around the world.
The Impact on Billions of People
UN researchers emphasize that environmental degradation associated with AI infrastructure does not affect all populations equally.
Communities most vulnerable to resource shortages often include:
- Rural populations
- Small farmers
- Indigenous communities
- Low-income households
- Drought-prone regions
- Climate-vulnerable countries
When water becomes scarce, land is degraded, or climate impacts intensify, the consequences are frequently felt first by populations with the fewest resources to adapt.
This is why many experts view AI sustainability as not only an environmental issue but also a question of global equity and development.
Industry Responds
Technology companies acknowledge environmental concerns but argue that AI can also be part of the solution.
Many firms are investing in:
- Renewable energy
- Water-efficient cooling systems
- Recycled-water usage
- Carbon-free electricity procurement
- Sustainable data-centre design
- AI-powered energy optimization
Industry leaders also argue that AI can help address environmental challenges by improving:
- Climate modelling
- Water management
- Precision agriculture
- Energy efficiency
- Biodiversity monitoring
- Disaster forecasting
According to this perspective, AI’s environmental costs must be weighed against its potential environmental benefits.
Can AI Become Sustainable?
Experts increasingly believe that sustainable AI will require coordinated action among governments, technology companies, researchers, and regulators.
Recommended measures include:
Greater Transparency
Mandatory reporting of energy, water, and land-use impacts.
Renewable-Powered Infrastructure
Expanding clean-energy deployment to meet growing AI demand.
Water-Efficient Cooling
Reducing freshwater dependence through advanced cooling technologies.
Responsible Supply Chains
Ensuring environmentally sustainable sourcing of minerals and hardware components.
Circular Economy Approaches
Increasing recycling and reuse of electronic equipment.
Environmental Standards
Establishing global sustainability benchmarks for AI infrastructure.
The Road Ahead
Artificial intelligence promises extraordinary benefits for humanity. It has the potential to accelerate scientific discovery, improve healthcare outcomes, increase productivity, and support economic growth.
However, the warnings from UN scientists highlight a critical reality: AI does not exist solely in the digital realm. It relies on physical systems that consume energy, water, land, and raw materials.
As governments and businesses race to expand AI capabilities, ensuring that growth remains environmentally sustainable will become one of the defining policy challenges of the coming decades.
The AI revolution is reshaping the global economy, but it is also reshaping humanity’s relationship with natural resources. Rising emissions, growing water consumption, expanding land requirements, and increasing demand for critical minerals are placing new pressures on ecosystems already strained by climate change and population growth.
The warning from UN researchers is not a call to halt technological progress. Rather, it is a reminder that innovation and sustainability must advance together. If artificial intelligence is to serve billions of people in the future, its development must be aligned with the protection of the natural resources upon which those billions ultimately depend.
UN researchers warn that artificial intelligence is increasing emissions, freshwater consumption, and land use, creating new environmental challenges that could affect billions worldwide.
