Report: AI Spurs 48% Rise in Google Emissions

0:00
/1861

Facts

  • In its environmental report published Tuesday, Google revealed that its greenhouse emissions in 2023 were 48% higher than in 2019 because its data centers had been using more power to meet the artificial intelligence (AI) surge.1
  • Google's emissions jumped 13% year over year in 2023, while its total data center electricity consumption grew 17% last year.2
  • The tech giant had planned to achieve net-zero emissions by 2030 but now admits that the 'greater intensity of AI comput[ing]' may make meeting that goal 'challenging.'3
  • This comes after Microsoft reported in May that its emissions had risen nearly 30% since 2020, primarily due to the construction of data centers.4
  • According to the International Energy Agency, data centers may use 1K terawatt hours of power annually by 2026 — nearly Japan's total electricity demand. By 2030, they are expected to use 4.5% of the world's electricity.5

Sources: 1Verge, 2CNBC, 3Bloomberg, 4Ft and 5Guardian.

Narratives

  • Narrative A, as provided by Mongabay Environmental News. The rise of AI and data centers threatens the environment with catastrophic effects. AI is driving water shortages, depleting scarce resources, and using vast amounts of electricity. Despite Big Tech's claim of efficiency, operations of companies like Google often prioritize profits over their severe socio-environmental consequences. The unchecked expansion of facilities like data centers even threatens to spark conflicts over essential resources.
  • Narrative B, as provided by Sustainability. Despite hindrances in achieving its 2030 goals, Google is working toward running its offices and data centers on 24/7 carbon-free energy, including solar and wind. Efforts include low-carbon building designs, energy efficiency, and electrification of facilities and fleets. Notably, the company's data centers are 1.8 times more energy efficient than most others. Google is also committed to sustainable commuting options and minimizing suppliers' emissions.

Predictions