Podcast about the subject Listen on Spotify
The expansion of artificial intelligence requires unprecedented amounts of water and energy.
The rapid advancement of artificial intelligence ushers in a historic turning point for humanity, opening unprecedented opportunities. But this transformation brings with it an environmental challenge that few discuss: the massive consumption of energy and water in the data centers that power these systems.
According to a report by Jornal Nacional, the operation of AI servers generates so much heat that data centers need sophisticated systems just for cooling. In addition to large-scale air conditioning, they use pipes with cold water that passes through the computers and is discarded heated. The paradox is that the biggest energy expenditure in cutting-edge technological development in Silicon Valley is on air conditioning..
How does energy consumption work?
Christopher Wellise, vice president of sustainability at Equinix, explains the challenge: the chips heat up intensely when electrical power passes through them, requiring constant heat removal. For an industrial-scale data center system, this demands immeasurable investments in cooling infrastructure.
Technology companies are signing long-term contracts with renewable energy suppliers to meet this growing demand. However, all the energy ends up mixed in the grid, making it difficult to truly trace green energy consumption.
Impact on climate goals
The issue gains urgency in a tense geopolitical context. The Donald Trump administration ordered a halt to new wind and solar energy projects, creating uncertainty about the ability of technology companies to continue generating renewable energy in sufficient volume to support the expansion of AI.
The dilemma is clear: the AI industry promises to revolutionize the economy and solve complex problems, but its current infrastructure jeopardizes the climate neutrality goals that countries like Brazil have set for the coming decades.
Photo by Growtika on Unsplash
English