Google's Electricity Consumption Skyrockets Amidst AI Expansion

Data analyst Kentan Joshi's analysis of Google's environmental report reveals a significant shift to exponential energy growth, raising concerns about the climate impact of generative AI.

Jul 2, 2026
Industry & Business
Google's Electricity Consumption Skyrockets Amidst AI Expansion

Google's electricity consumption has seen a dramatic increase, shifting from linear to exponential growth over the past two years, according to an analysis by data analyst Kentan Joshi.

Joshi, who examined Google's latest environmental report, noted that the company's total electricity consumption rose from 31 terawatt hours (TWh) in 2024 to 43 TWh in 2025. This represents the largest single-year increase in the company's power consumption on record. "A wild testament to the obscene bloat and waste of GenAI," Joshi stated on the social platform BlueSky.

This surge in energy demand places Google's power usage on par with entire countries. Last year, the company's consumption exceeded that of nations such as Slovakia, Ecuador, Ireland, and Nigeria.

While Google utilizes renewable energy sources, Joshi's analysis suggests this may not be sufficient to offset the increased consumption. Even after adjustments for supply chain exclusions and renewable energy procurement claims, Google's greenhouse gas emissions have reportedly increased by nearly 50% in the last five years, moving counter to its 2021 emissions targets.

Joshi's critique, citing Google's own environmental report, questions the company's rapid AI infrastructure buildout when the grid's decarbonization pace is slower. He posits that if operations are pushing planetary boundaries, Google should re-evaluate the worth of its AI endeavors.

The analysis also addresses common misconceptions about the energy cost of AI. Joshi explains that text generation by chatbots represents a comparatively low-energy form of generative systems, but AI encompasses a much broader range of technologies with significantly higher energy demands. He previously argued that industry-wide metrics, such as claims of minimal energy per prompt, can obscure the true environmental impact of the broader AI sector. Joshi characterizes this selective disclosure as a strategy to control the narrative around AI's environmental footprint.

Sources