The rapid expansion of artificial intelligence data centers built by Big Tech companies — including Microsoft, Google and Amazon — is having an enormous environmental impact. However, according to a Mongabay report, some regions are feeling the effects more than others.
What is going on?
Mongabay explained that AI data centers are much larger and use more resources than their recent predecessors. To keep up with the AI explosion, each facility contains thousands of servers, miles of cables, and massive computing loads.
“A single AI campus can be bigger than a small town,” Mongabay reported.
The energy demands of these operations are staggering. Global data center electricity use is forecast to reach about 536 terawatt-hours this year, according to the outlet. This is about 2% of global energy use. As AI workloads accelerate, that figure could double by 2030.
All that energy produces intense heat, forcing the centers to rely on extensive cooling systems – many of which rely heavily on fresh water. By 2027, AI centers may need up to 6.4 trillion liters of water per year. That’s the equivalent of nearly 2.8 million Olympic-sized pools.
In addition to energy and water pressure, data centers consume tremendous quantities of raw materials. Copper is particularly critical for wiring and other key components. Mongabay reported that it may be necessary to mine as much copper in the next 25 years as has been mined in all of history to meet demand. Meanwhile, servers and components are quickly replaced, generating significant e-waste. With less than 25% of global e-waste properly recycled, most of it poses long-term pollution risks.
However, all this development of taxation is not happening in the rich countries where AI is used the most. Instead, low-income countries in the Global South are being overtaken by massive data centers because they offer cheap land and labor, weak environmental regulations, and abundant natural resources that Big Tech can exploit with limited oversight.
Critics argue that this rapid expansion of AI is “extractive colonialism,” in which wealthy Northern tech companies deploy resource-intensive infrastructure in poorer regions of the Global South. This is often with little regard for the water, energy or environmental needs of local communities.
“This kind of Big Tech exploitation … seems to echo older forms of extractive colonialism seen in the tobacco, cotton, coffee and sugar plantations that once fed the tastes and vices of the Global North,” Mongabay reported.
Why is this important?
Simply put, AI users are often removed from the environmental and human burden of the growing technology. Understanding the true impact of AI is essential to creating a more sustainable — and more responsible — future for the sector.
“We really need to talk about what is an appropriate level of technological progress,” Bryan Bixcul, global coordinator at the SIRGE Coalition, told Mongabay.
Internet users are rapidly adopting AI for everything from simple entertainment to professional work — and that boom is leading to an increase in energy- and resource-intensive data centers. As these facilities burn through enormous amounts of resources, critics warn that their growing footprint could undermine global efforts to reduce carbon pollution.
As global temperatures rise due to climate change, many low-income tropical regions are already experiencing water insecurity due to drought. With data centers entering the picture, some communities – such as the residents of the state of Querétaro in Mexico – are concerned that their already depleted resources will go towards cooling the data centers rather than supporting local communities.
“Water is what is needed for the people, not for these industries,” Querétaro campaigner Teresa Roldán told the BBC.
AI data centers are also driving additional demand for dirty energy sources like oil and coal, complicating global climate goals and slowing the transition to cleaner energy. Currently, dirty energy sources provide almost 60% of data center power. With AI adoption expanding rapidly, that number could rise to keep up with demand.
What is being done to monitor the expansion of AI?
The short answer: not much — especially on a global scale.
Most countries still lack clear policies governing where, how, or how fast data centers can expand. This means that Big Tech often makes decisions with little local oversight or input. In many low-income countries, governments even welcome these projects as a means of economic growth, despite the environmental trade-offs.
While researchers and nonprofits have begun to track and measure the environmental cost of AI infrastructure, there is still no global standard for reporting energy, water or material use — and most available data comes from voluntary corporate disclosures that are not independently verified. Basically, this means that tech companies are self-publishing sustainability reports that may (or may not) be completely transparent.
While some regions now require data center operators to report their energy and water consumption, these efforts are mostly concentrated in wealthier northern countries. As Mongabay pointed out, much of the fastest AI growth is occurring in regions with the weakest oversight.
Get TCD’s free newsletters for easy tips to save more, waste less, and make smarter choices — and earn up to $5,000 toward clean upgrades in TCD’s exclusive Rewards Club.