The global expansion of data centers linked to artificial intelligence opens a new environmental front. Consequently, a study by Cambridge University warns about the formation of “data heat islands”.
These increasingly large facilities release heat into the environment. Furthermore, their accelerated growth responds to the demand for AI processing.
According to the research, the average thermal increase reaches 2°C. However, in extreme cases, it can reach 9.1°C.
This phenomenon could impact more than 340 million people. Therefore, it becomes an emerging concern.
A global phenomenon with focuses in Europe and Latin America
The analysis used NASA satellite data over two decades. Consequently, patterns of warming associated with more than 6,000 data centers were identified. In Aragón, an increase of 2°C was recorded. Likewise, this value exceeds that of nearby regions.
A similar behavior was observed in the Bajío. There, the thermal increase coincides with technological expansion. In Ceará and Piauí, the increase reached 2.8°C. In particular, the city of Teresina showed unusual values.
Moreover, the thermal effect extends up to 10 kilometers. Thus, it replicates the urban heat island phenomenon.
The energy consumption behind digital growth
The growth of artificial intelligence demands enormous volumes of energy. Firstly, data centers require constant electricity.
Additionally, cooling systems increase energy consumption. Consequently, much of the energy dissipates as heat.
It is estimated that, in five years, data processing could surpass the energy consumption of global manufacturing. Therefore, the sector positions itself as one of the most demanding.
Likewise, this intensive consumption poses challenges for the energy transition. Thus, it increases the pressure on electrical grids.
Environmental impact and regulatory challenges
Localized warming alters ecosystems and living conditions. Consequently, it affects both biodiversity and human communities.
Moreover, the lack of regulation exacerbates the problem. Therefore, technological development advances without sufficient controls.
Unlike other studies focused on emissions, this analysis addresses physical heat. Thus, it broadens the understanding of the environmental impact.
However, some specialists believe that more evaluations are required. In this way, the aim is to accurately measure the phenomenon.
Technological solutions to warming
Among the proposed solutions, efficient software stands out. Consequently, it would allow reducing the energy consumption of AI training. Additionally, hardware capable of reusing energy is being researched. Therefore, it would decrease heat dissipation.
Likewise, technologies such as radiative cooling offer sustainable alternatives. Thus, they could reduce consumption by between 8% and 20%.
Improving territorial planning is also proposed. Thus, it would prevent the concentration of infrastructure in sensitive areas.
An urgent balance between innovation and sustainability
The advancement of artificial intelligence redefines the global economy. However, it also generates new environmental impacts.
Consequently, there is a need to integrate ecological criteria. Additionally, greater supervision is required. The challenge is to decouple technological progress from environmental damage. Therefore, it becomes crucial to rethink the model.
Ultimately, “data heat islands” highlight a growing tension. Thus, the digital future will depend on its sustainability.
