Google and Amazon's green goals falter as AI costs mount
Google and Amazon reported significant increases in carbon emissions, indirectly driven by the soaring energy demands of artificial intelligence. Emissions rose 25% and 16% respectively last year, putting their net-zero pledges at risk.
Big Tech's race to adopt artificial intelligence is clashing with its climate ambitions. New sustainability reports from Google and Amazon reveal that both companies saw substantial jumps in their overall carbon emissions last year: Google by 25% and Amazon by 16%. While neither explicitly blames AI, the reports offer ample circumstantial evidence linking the growth to AI infrastructure.
The main culprit is Scope 3 emissions—indirect pollution from purchased goods, services, and supply chains. For Google, Scope 3 emissions rose by 2.1 million metric tons last year, doubling from its 2019 baseline. Amazon's Scope 3 increase stems largely from capital goods, including data centers and warehouses. The company noted it added more data center capacity than any other in 2025, with 1.2 gigawatts in Q4 alone. These facilities require vast amounts of steel, cement, and semiconductors—all heavy emitters during production.
Historically, both companies offset their operational energy use through renewable energy certificates. However, AI's relentless power needs are pushing them back toward fossil fuels. Semiconductor manufacturing, concentrated in Asia where grids rely on coal and gas, is especially carbon-intensive. The chemicals used in chip fabrication are also potent greenhouse gases.
Achieving net-zero goals now demands even more aggressive action: expanding renewable procurement, investing in low-carbon steel and cement, and purchasing millions of tons of carbon removal credits. AI has made the path steeper, though not impossible. As TechCrunch's analysis suggests, the numbers in these reports serve as a warning sign of AI's real environmental cost.
