With AI energy demand surging Google reports jump in emissions

1 min read

Google has reported an almost 50 per cent increase in emissions over the past five years, which suggests that its goal of reducing its climate footprint is now at risk.

Credit: Beenish - adobe.stock.com

With the company increasingly reliant on data centres, which are extremely energy-hungry, to power its new artificial intelligence products its greenhouse gas emissions have jumped by a startling 48% since 2019.

According to Google, its data centres are driving the company’s demand for electricity and in its annual environmental report it revealed that emissions in 2023 had risen 13% compared with the previous year, reaching a total of 14.3m metric tons.

So how do companies like Google square the challenge of not only investing heavily in AI but in reaching net zero by 2030, which is the company’s stated aim.

Well, it fudges it! According to Google there is now “significant uncertainty” around reaching the target. There’s also much greater, “uncertainty around the future environmental impact of AI,” which is both “complex and difficult to predict”.

The International Energy Agency estimates that data centres’ total electricity consumption could double from 2022 levels to 1,000TWh (terawatt hours) in 2026, and research suggests that AI will result in data centres accounting for some 4.5% of global energy generation by 2030.

Microsoft has also admitted that energy use related to its data centres was endangering its “moonshot” target of being carbon negative by 2030 due to its AI strategy.

There have been plenty of pledges to reduce CO2 emissions, but now these are now coming up against other pledges to invest heavily in AI products.