Google’s AI Drives Sharp Rise in Greenhouse Gas Emissions

As Google has quickly integrated artificial intelligence into its key products, the company has faced a significant challenge: the AI systems have drastically increased Google’s greenhouse gas emissions.

These AI systems require vast amounts of computing power. The data centers that operate them, essentially large facilities filled with powerful computing equipment, consume enormous amounts of energy to process data and manage the heat generated by the computers.

As a result, Google’s greenhouse gas emissions have surged by 48% since 2019, according to the company’s annual environmental report. Google attributes this rise primarily to increased energy consumption in data centers and supply chain emissions.

Google now describes its goal of achieving net-zero emissions by 2030 as “extremely ambitious.” The company acknowledges that this goal may be influenced by “the uncertainty around the future environmental impact of AI, which is complex and difficult to predict.” In other words, Google’s sustainability efforts — once encapsulated by the motto “don’t be evil” — have become more complicated due to AI.

Like its tech industry peers, Google has heavily invested in AI, viewing it as the next significant technological revolution that will transform our lives, work, and information consumption. The company has integrated its Gemini generative AI technology into core products like Search and Google Assistant, with CEO Sundar Pichai declaring Google an “AI-first company.”

However, AI’s major drawback is its substantial energy requirement. Data centers, which Google and other tech giants are expanding with massive investments each quarter, consume vast amounts of power to support AI ambitions. The International Energy Agency estimates that a typical Google search query uses 0.3 watt-hours of electricity, whereas a ChatGPT request uses about 2.9 watt-hours. Dutch researcher Alex de Vries’s October study suggested that Google’s AI systems could potentially consume as much electricity annually as the entire country of Ireland, assuming full-scale adoption with current hardware and software.

“As we further integrate AI into our products, reducing emissions may be challenging due to increasing energy demands from the greater intensity of AI compute, and the emissions associated with the expected increases in our technical infrastructure investment,” Google stated in its report published Monday. The company added that the electricity consumption of its data centers is currently rising faster than its ability to bring carbon-free energy sources online.

Google anticipates its total greenhouse gas emissions will continue to rise before eventually declining, as it invests in clean energy sources like wind and geothermal to power its data centers.

The significant water usage for cooling to prevent data centers from overheating also poses a sustainability challenge. Google aims to replenish 120% of the freshwater it uses in its offices and data centers by 2030. Last year, it replenished just 18%, although this was a significant increase from 6% the previous year.

Additionally, Google is exploring ways to use AI to combat climate change. A 2019 project by Google DeepMind trained an AI model on weather forecasts and historical wind turbine data to predict wind power availability, thereby enhancing the renewable energy source’s value for wind farmers. Google has also employed AI to suggest more fuel-efficient routes for drivers using Google Maps.

“We know that scaling AI and using it to accelerate climate action is just as crucial as addressing the environmental impact associated with it,” Google stated in the report.