Google has released its 2024 Environmental Report, a more than 80-page document outlining all of the giant company’s efforts to apply technology to environmental issues and mitigate its own contribution. But he completely avoids the question of how much energy AI uses – perhaps because the answer is “a lot more than we’d like to say”.
You can read the full report here (PDF), and it honestly has a lot of interesting stuff in it. It’s easy to forget just how many plates a company as big as Google keeps spinning, and there’s some truly remarkable work in here.
For example, he works in one water replenishment program, with which it hopes to offset the water used in its facilities and operations, ultimately creating a net positive. This is done by identifying and funding watershed restoration, irrigation management and other work in this area, with dozens of such projects around the world being at least partially funded by Google. It has reached 18% of water use being replenished (by whatever definition of the word is used here) in this way and is improving every year.
The company is also keen to tout AI’s potential climate benefits, such as optimizing irrigation systems, creating more efficient routes for cars and boats, and predicting floods. We’ve highlighted some of these already in our AI coverage, and they could actually be quite useful in many areas. Google doesn’t have to do these things, and many big companies don’t. Well, credit where credit is due.
But then we get to the “Responsible management of AI resource consumption” section. Here Google, so confident in every statistic and estimate up until now, suddenly throws up its hands and shrugs. How much energy does artificial intelligence use? Can anybody Really be sure;
But it must be bad, because the first thing the company does is downplay the entire data center energy market, saying it’s only 1.3% of global energy use, and the amount of energy Google uses is only 10% at most — so only 0.1% of the world’s total energy powers its servers, according to the report. A trifle!
In particular, it decided in 2021 that it wanted to reach net zero emissions by 2030, although the company admits there is a lot of “uncertainty,” as it likes to call it, about how that will actually happen. Especially since its emissions have been increasing every year since 2020.
In 2023, the total of our greenhouse gases [greenhouse gas] emissions were 14.3 million tCO2er, representing a 13% year-over-year increase and a 48% increase compared to our 2019 target base year. This result was mainly due to increases in data center energy consumption and supply chain emissions. As we further integrate artificial intelligence into our products, reducing emissions may be difficult due to the increasing energy demands from the greater computation intensity of artificial intelligence and emissions associated with expected increases in our technical infrastructure investments.
(Put my emphasis on this and the quote below.)
However, the development of artificial intelligence is lost among the aforementioned uncertainties. Google has the following justification for why the company is not specific about the contribution of AI workloads to the overall data center energy bill:
Predicting the future environmental impacts of AI is complex and evolving, and our historical trends likely do not fully capture the future trajectory of AI. As we integrate AI deeply into our product portfolio, the distinction between AI and other workloads will be meaningless. So we focus on metrics across the data center since they include the overall resource consumption (and therefore environmental impact) of AI.
“Complex and evolving” “Trends likely not fully captured.” “the distinction… won’t make sense”: This is the kind of language used when someone knows something but would really, really prefer not to tell you.
Does anyone actually believe that Google doesn’t know, down to the last penny, how much training and inference they have added to their energy costs? Isn’t analyzing this data so accurately part of the company’s core competency in cloud computing and data center management? It has all these other statements about how efficient custom AI server modules are, how it does all this work to reduce the energy required to train an AI model by 100 times and so on.
I have no doubt that there are many great green efforts going on at Google and you can read all about it in the report. But it’s important to point out what he seemingly denies: the huge and growing energy costs of AI systems. The company may not be the main driver of global warming, but despite its potential, Google doesn’t seem to be in a net positive position just yet.
Google has every incentive to downgrade and obscure these elements, which even in their reduced, highly efficient state, can hardly be good. We’ll definitely be asking Google to get more specific before we find out if they get even worse in the 2025 report.
