Microsoft, already one of the biggest polluters because of AI, saw its 2023 emissions grow 30% from 2020

The rapid expansion of artificial intelligence (AI) is reshaping the tech and energy sectors, but it’s also driving up emissions and water usage, raising alarms about environmental consequences.

Microsoft recently disclosed that its emissions in 2023 surged by 29.1% compared to 2020 levels, mainly due to the construction of new data centres and the carbon footprint of building materials and hardware. This jump illustrates how the pursuit of AI is intersecting with efforts to reduce emissions.

To mitigate its environmental impact, Microsoft is ramping up efforts to use greener materials like steel, concrete, and fuels. Additionally, the company aims to power all its corporate operations with renewable energy by 2025 and achieve zero-carbon energy sources 24/7 by 2030.

However, the growth of data centres for AI and cloud computing could drive up the demand for natural gas, posing challenges for reducing emissions. This growth also presents opportunities for pipeline companies like Energy Transfer and Kinder Morgan, as highlighted by East Daley Analytics.

Microsoft is also exploring next-generation nuclear reactors to power its data centers and AI initiatives, aiming to address the substantial electricity consumption associated with these operations. While nuclear energy doesn’t emit greenhouse gases, it brings concerns about managing radioactive waste and establishing a secure uranium supply chain.

For its star data centre, the one it is developing with OpenAI and will be investing $100 billion in, Microsoft is planning to power it using an in-house nuclear reactor.

The escalating water consumption by tech giants like Microsoft, Google, and Meta for cooling their data centres has also raised environmental concerns. As these companies expand their AI capabilities, the demand for computing power grows, leading to massive server farms sprouting up across the world. These server farms rely on water systems that pump chilled water through the servers, a process which again, consumes a lot of water and electricity.

A study from the University of California, Riverside emphasizes the urgent need to address the undisclosed water footprint of AI models, especially amid increasing freshwater scarcity and ageing water infrastructure. Microsoft alone consumed approximately 22 billion litres of water in 2022, with Google and Meta also contributing significant amounts.

In response to these concerns, tech companies have set targets to replenish water resources by 2030, aiming to return more water than they consume through initiatives like enhancing irrigation infrastructure and restoring wetland systems.

As AI continues to reshape industries, the environmental impact of its growth requires careful consideration and proactive measures to ensure sustainability. Balancing technological advancements with environmental responsibility is crucial for building a more sustainable future.