"Side effects" of AI applications: Google's carbon emissions surge nearly 50% in five years

Google's recent data revealed the "side effects" of the growth of artificial intelligence (AI) applications on Silicon Valley giants: over the past five years, Google's greenhouse gas emissions have surged by 48% due to the expansion of data centers supporting AI systems.

On Tuesday, July 2nd, Eastern Time, Google's annual environmental report showed that in 2023, last year, Google's pollution emissions reached 14.3 million tons of carbon equivalent, a 48% increase from the 2019 baseline level, with a 13% increase over the past year. Last year, Google's energy-related emissions from data center electricity consumption increased by 37%, accounting for a quarter of the company's total greenhouse gas emissions, with the largest share of 75% of supply chain emissions increasing by 8% last year.

The report highlights that last year's increase in carbon emissions underscores the challenge of reducing emissions while investing in large language models (LLMs) and related applications and infrastructure construction. The report acknowledges that the "future impact of AI on the environment" is "complex and difficult to predict." The report predicts that supply chain emissions will continue to grow in the near future due to the increasing infrastructure requirements for running AI systems.

The report on Tuesday implies that the electricity demand generated by AI systems poses a threat to Google's ability to achieve its decarbonization goals. Google had previously committed to achieving net zero direct and indirect greenhouse gas emissions by 2030, with its grid running on zero-carbon energy sources 24 hours a day by then.

However, the report on Tuesday warned that last year, some clean energy projects were "terminated," reducing the amount of renewable energy Google was able to obtain.

Kate Brandt, Google's Chief Sustainability Officer, stated in the report that Google remains committed to achieving its 2030 goals, but emphasized that the goal is "extremely ambitious." Brandt said, "We do expect our emissions to continue to rise before they fall back to our target." She also mentioned that Google is "working very hard" to reduce emissions, including signing clean energy agreements, and AI provides a "huge opportunity" to address climate issues.

Media reports have pointed out that Silicon Valley giants like Google, Amazon, and Microsoft have previously announced plans to invest billions of dollars in the AI field, raising concerns among climate experts about the negative environmental impact of energy-intensive tools and systems.

Wall Street News mentioned earlier this week that with the rapid development of AI technology, especially the rise of LLM and generative AI, technology companies' demand for electricity is growing explosively, leading them to compete for nuclear power resources. About one-third of U.S. nuclear power plant owners are in talks with technology companies to provide power for new data centers. Among them, Amazon Web Services (AWS) is close to reaching a direct power supply agreement with Constellation Energy, the largest nuclear power plant operator in the United States. In March of this year, AWS also acquired a nuclear-powered data center in Pennsylvania for $650 million This nuclear power - technology giant alliance has sparked controversy in many parts of the United States. On the one hand, the combination of nuclear power and data centers can match the most reliable electricity in the grid with the wealthiest customers. Moreover, this new "factory-grid integration" arrangement means that the construction speed of data centers may be greatly accelerated, as there is almost no need for new grid infrastructure. Data centers may also avoid paying the hefty transmission and distribution fees that dominate electricity bills.

On the other hand, such power supply agreements also mean that technology companies are not meeting their soaring power needs by increasing new green energy, but are actually transferring existing power resources. Data centers may draw stable generation resources from the grid, exacerbating the increasingly severe power reliability issues in many parts of the United States and driving up electricity prices