AI, Data Centres and the Urban Food Challenge: What Communities Are Saying About the Digital Future
What happens when the “cloud” needs land, water and electricity?
Artificial intelligence often feels invisible. We ask a chatbot a question, generate an image, translate a sentence or use an AI-powered service, and the result appears almost instantly.
But behind that apparently weightless digital experience is a very physical infrastructure: data centres filled with servers, cooling systems, electricity networks, water supplies, land and construction materials.
And this creates a question that is increasingly being asked by communities around the world:
Who gets the benefits of AI—and who carries the environmental and social costs?
This question becomes particularly important when data centres are built close to cities, agricultural areas or already water-stressed communities. In India, researchers and civil-society organisations are increasingly examining how data-centre development affects land, water and livelihoods. A 2026 Digital Empowerment Foundation study of peri-urban Telangana, for example, highlights concerns around land loss, water access, changing livelihoods and the need for greater community participation in decisions about digital infrastructure.
At the same time, cities are already struggling with another basic question: How can enough affordable, nutritious food reach growing urban populations while water, energy and land become increasingly constrained?
This is where AI, data centres and urban food systems become connected.
1. AI Is Digital—but Its Infrastructure Is Physical
Artificial intelligence depends on enormous amounts of computing power.
Training and running advanced AI systems requires:
- computer servers
- specialised chips
- electricity
- cooling equipment
- buildings
- water or other cooling resources
- telecommunications networks
- land
- minerals and manufacturing supply chains.
The International Monetary Fund has described this physical foundation as an AI resource race involving energy, chips and minerals. Large data centres can have power requirements measured in tens of megawatts, while newer projects are moving toward much larger scales.
This changes how we should think about AI.
AI is not simply:
software → information → convenience
It is also:
AI → data centres → electricity + water + land + infrastructure → communities
That second chain is often less visible to users.
2. Why Communities Are Becoming Central to the AI Debate
When a major data centre is proposed, the discussion often focuses on economic benefits.
Developers and governments may highlight:
- employment
- investment
- tax revenue
- infrastructure
- digital development
- technological leadership
- local economic growth.
These benefits can be significant.
However, communities may ask different questions:
- How much electricity will the facility consume?
- Where will that electricity come from?
- How much water will cooling require?
- Will household water supplies be affected?
- Will electricity prices increase?
- What happens to agricultural land?
- Will farmers lose access to groundwater?
- What noise will the facility generate?
- How many permanent local jobs will actually be created?
- Who gets to make these decisions?
Recent research on community responses to data-centre development identifies recurring concerns around environmental risk, infrastructure pressure and lack of transparency. It also finds that early engagement and measurable community benefits can improve trust.
3. Water May Become the Most Important Community Question
Data centres generate considerable heat.
Cooling systems are therefore essential.
Depending on the technology and local conditions, data-centre cooling can involve substantial water consumption. Water is also used indirectly because electricity generation itself can have a water footprint.
A 2025 study in Nature Sustainability estimated that AI-server deployment in the United States could produce an annual water footprint of 731–1,125 million cubic metres between 2024 and 2030, depending on the scale and location of deployment.
Another analysis found that many newly developed data centres are located in areas already experiencing significant water stress.
The community issue is therefore not simply:
“Does the data centre use water?”
It is:
“Whose water is being used, and what happens when different users compete for the same resource?”
4. India: Data Centres and Water-Stressed Regions
This question is particularly relevant to India.
India's data-centre industry is expanding rapidly as cloud computing, digital services and AI adoption grow.
Research from MIT's Abdul Latif Jameel Water and Food Systems Lab notes that India's data-centre capacity is expected to grow substantially, with many planned facilities located in areas where groundwater is already important for agriculture. The research specifically examines how data-centre development can compete with agricultural and urban groundwater needs.
This creates a food-water-energy nexus.
A simplified picture looks like this:
Data centre
↓
Electricity demand
↓
Water and energy infrastructure
↓
Competition with households / agriculture / ecosystems
↓
Possible consequences for food production and livelihoods
The key point is that these systems cannot always be planned separately.
5. What Does This Have to Do With Urban Food Systems?
At first, AI data centres and urban food systems might seem like completely different topics.
They are not.
An urban food system includes the whole chain through which food reaches people in cities:
production → processing → transport → markets → retail → consumption → waste
Urban food systems depend on:
- land
- water
- energy
- transport
- labour
- markets
- infrastructure
- waste-management systems.
A 2025 study using the food-energy-water nexus approach argues that urban food ecosystems face interconnected pressures involving energy use, water scarcity, food waste and inefficient distribution. These pressures can contribute to environmental degradation and unequal access to nutritious food.
So when a new piece of infrastructure competes for scarce land, water or electricity, it can indirectly affect the food system.
6. The Food–Water–Energy Nexus
One of the most useful concepts for understanding this topic is the Food–Water–Energy Nexus.
These three systems are interconnected.
Food needs water
Agriculture requires water for irrigation, livestock and processing.
Food needs energy
Energy is required for:
- pumping
- refrigeration
- processing
- transport
- storage
- retail.
Energy needs water
Some forms of electricity generation require water.
Data centres need energy and cooling
AI infrastructure therefore enters the same resource system.
The result is a complex network:
Water → agriculture → food
Water → cooling → data centres
Energy → farming and food distribution
Energy → AI and data centres
This is why community planning needs to consider the whole system rather than one project at a time.
7. Land Is Another Major Issue
Data centres require large physical sites.
When they are constructed on the urban fringe or peri-urban areas, they may compete with:
- farmland
- grazing areas
- wetlands
- forests
- residential land
- community spaces.
The Digital Empowerment Foundation's research in Telangana specifically examines how data-centre expansion is changing land use and livelihoods in peri-urban communities.
This is particularly significant because peri-urban land often performs multiple functions.
The same landscape may provide:
food + income + housing + water recharge + cultural identity + ecological services
Converting it to industrial infrastructure can therefore produce effects that are much larger than the loss of land measured in hectares.
8. The Telangana Example
The “Just AI, Just Land” research provides an important Indian community perspective.
The study examines data-centre development in peri-urban areas of Telangana, including Mekaguda and Begarikancha.
It looks beyond the technological promise of AI and asks what digital infrastructure means for people living near it.
The research identifies changes involving:
- land use
- water access
- agricultural livelihoods
- local economic opportunities
- relationships with natural resources
- local governance.
Most importantly, it argues for stronger:
- information sharing
- transparency
- community participation
- local-level decision-making.
This illustrates a broader principle:
AI governance cannot stop at algorithms.
It must also consider the places and people affected by the infrastructure that makes AI possible.
9. Farmers and Food Producers Need a Voice
When data-centre development takes place in agricultural regions, farmers become important stakeholders.
Their concerns may include:
- groundwater availability
- irrigation costs
- land acquisition
- changes in land prices
- access to roads
- electricity availability
- changing employment patterns.
The question is not necessarily whether technology and agriculture must compete.
Instead, planning should ask:
Can digital infrastructure be located and designed in ways that minimise competition with essential agricultural resources?
Possible strategies include:
- avoiding severely water-stressed agricultural areas
- using treated wastewater
- improving cooling efficiency
- monitoring groundwater
- establishing water-use limits
- conducting cumulative environmental assessments
- involving farmers in planning.
MIT researchers are exploring precisely these kinds of location and policy questions for data centres in water-stressed parts of India.
10. AI Can Also Help Urban Food Systems
The relationship isn't entirely negative.
AI can potentially improve urban food systems.
For example, AI and data analytics can assist with:
Crop monitoring
Satellite imagery and machine learning can identify crop stress and water requirements.
Demand forecasting
Retailers and food distributors can predict demand more accurately and potentially reduce food waste.
Logistics
AI can optimise delivery routes and distribution networks.
Irrigation
Data-driven systems can help farmers apply water more precisely.
Food waste management
AI can identify patterns in food waste and improve inventory management.
Urban agriculture
Sensors and predictive models can help monitor temperature, humidity, nutrients and water use.
So the challenge is not:
AI vs. food security
It is:
How can AI contribute to food security without creating new resource pressures?
11. The Risk of “Efficiency” Without Equity
A technology can become more efficient while the overall system becomes less equitable.
For example, suppose AI makes a logistics system more efficient.
That sounds positive.
But if the infrastructure supporting that AI:
- consumes scarce groundwater,
- raises local electricity demand,
- occupies agricultural land,
- or shifts environmental costs onto poorer communities,
then efficiency at one level may create inequality at another.
This is why efficiency alone is not enough.
A sustainable system must ask:
Efficient for whom?
At whose expense?
Who decides?
Who benefits?
12. Environmental Justice
The idea of environmental justice is central to community perspectives on AI infrastructure.
Environmental justice asks whether environmental benefits and burdens are distributed fairly.
For data centres, this means examining:
Benefits
- jobs
- investment
- infrastructure
- tax revenue
- digital services.
Potential burdens
- water consumption
- noise
- air pollution
- land-use change
- energy demand
- pressure on infrastructure
- ecological impacts.
Research on communities around data centres has documented concerns involving noise, water, air pollution, infrastructure and household economic pressures.
The critical question is therefore:
Do the communities hosting AI infrastructure receive a fair share of its benefits?
13. Transparency Matters
One recurring issue is information.
Communities cannot participate meaningfully if they do not know:
- how much water a facility will use,
- how much electricity it requires,
- where electricity will come from,
- what land is being acquired,
- how many jobs will be created,
- what environmental safeguards exist.
Research on data-centre water governance has argued that public access to water-use information is important for democratic decision-making.
Therefore:
Transparency is not just public relations.
It is a condition for meaningful participation.
14. From Consultation to Participation
There is a major difference between:
Consultation
and
Participation.
Consultation
A developer tells the community:
“Here is what we plan to do. What do you think?”
Participation
The community has meaningful opportunities to influence:
- site selection
- environmental safeguards
- water arrangements
- infrastructure planning
- benefit-sharing
- monitoring
- future expansion.
Community-centred research increasingly points toward early engagement, transparent commitments and mechanisms such as community advisory boards or community-benefit agreements.
15. What Could a Community-Centred AI Model Look Like?
A more responsible approach could include seven principles.
1. Put communities first
Residents should be involved before major decisions are finalised.
2. Measure resource use
Water and electricity consumption should be transparent and independently verifiable.
3. Protect essential resources
Household water and agricultural needs should not be displaced by industrial demand.
4. Choose sites carefully
Data centres should be located according to water availability, grid capacity, ecological sensitivity and land-use priorities—not simply cheap land.
5. Use cleaner cooling
Developers should invest in water-efficient or water-reuse cooling technologies where appropriate.
6. Share benefits
Communities should see tangible benefits through jobs, skills, infrastructure and local investment.
7. Monitor continuously
Community engagement should continue after construction rather than ending when permits are approved.
16. Urban Food Systems Need Similar Community Participation
The same principle applies to food.
Urban food policy should involve:
- farmers
- street vendors
- market traders
- consumers
- municipal authorities
- food businesses
- waste workers
- urban gardeners
- community organisations.
People who experience food insecurity often understand barriers that a purely technical model may overlook.
A participatory study of the Pune and Bhima Basin involving 75 resource users and experts identified interconnected pressures involving urbanisation, climate change, groundwater extraction, land-use change, water scarcity and inefficient infrastructure.
This demonstrates why local knowledge is valuable.
17. The Bigger Urban Question
Cities are increasingly being asked to accommodate several competing priorities:
Housing
Food
Water
Energy
Transport
Digital infrastructure
Green space
Industry
These cannot always be planned independently.
A city that attracts AI infrastructure but loses agricultural land and water security may gain digital capacity while becoming more vulnerable in other ways.
The goal should therefore be:
A resilient city, not simply a smarter city.
18. AI and the Future of Sustainable Cities
The future of urban development will probably involve much greater integration between AI and physical infrastructure.
AI may help cities:
- predict water demand
- manage electricity grids
- optimise public transport
- reduce food waste
- monitor pollution
- manage urban farms
- improve emergency response
- forecast climate risks.
But these benefits depend on infrastructure.
The paradox is:
AI can help cities become more efficient while simultaneously increasing the resource demand of the infrastructure that runs AI.
Recognising this paradox is essential to responsible urban planning.
19. The Central Community Perspective
From a community perspective, the most important issue is not whether AI is “good” or “bad.”
That is too simple.
The better questions are:
Who owns the infrastructure?
Who controls the resources?
Who makes the decisions?
Who receives the economic benefits?
Who experiences the environmental costs?
What happens to farmers and food producers?
What happens to water security?
What happens to future generations?
These questions shift the discussion from technological progress to socially accountable development.
20. Conclusion: Building a Digital Future That Feeds Its Cities
AI will continue to expand, and data centres will become an increasingly important part of urban and national infrastructure.
But digital development cannot be separated from the physical world.
Every AI system ultimately depends on:
land + energy + water + materials + people + communities.
At the same time, cities depend on those same resources to produce and distribute food.
This makes the connection between AI, data centres and urban food systems more than an environmental issue. It is a question of justice, governance and urban resilience.
A community-centred approach does not mean rejecting AI.
It means asking AI to fit within the needs and limits of the places where it is built.
The future should not simply be about creating smarter cities.
It should be about creating cities that are smarter, fairer, more resource-secure and capable of feeding their people.
That means putting communities—not just computation—at the centre of the digital future.
Tags:
AI, Artificial Intelligence, Data Centres, Urban Food Systems, Community Perspectives, Food Security, Water Security, Energy, Sustainable Cities, Digital Infrastructure, Environmental Justice, India, Urban Agriculture, Climate Change, Responsible AI