Google’s electricity consumption surged 37% in 2025 — the largest annual increase in the company’s history — driven almost entirely by the explosive growth of its AI data center operations, according to the company’s 2026 Environmental Report. The jump underscores the enormous and growing energy demands of the artificial intelligence industry at a moment when the world’s power grids are struggling to keep up.
Since 2019, Google’s electricity use has increased by more than 250%, a trajectory that has forced the company to confront uncomfortable questions about the environmental cost of the AI revolution it is helping to lead. The company has pledged to run its operations on carbon-free energy by 2030, but the sheer scale of its AI buildout is making that goal increasingly difficult to reach.
The Scale of the Problem
Google is not alone. Microsoft, Amazon, and Meta are all reporting similar surges in energy consumption as they race to build the data centers needed to train and run large AI models. A single query to an AI-powered search engine can use up to ten times more electricity than a traditional search, and training a frontier model like GPT-5 or Google’s Gemini can consume as much power as a small city over the course of months.
The International Energy Agency has warned that data center electricity consumption could double by 2030, driven primarily by AI. In some markets, utilities are already delaying the retirement of coal and natural gas plants to ensure they can meet the demand from new data centers — a development that complicates the tech industry’s climate commitments.
Google’s Response
Google has been one of the largest corporate purchasers of renewable energy for years, and the company says it matched 100% of its annual electricity consumption with renewable energy purchases. But critics note that these purchases — often in the form of renewable energy certificates — do not mean the company is actually running on clean energy in real time. The data centers are connected to the same grids as everyone else, and when those grids are powered by fossil fuels, so are the data centers.
The company has also invested in advanced geothermal energy, next-generation nuclear, and battery storage as part of its effort to decarbonize its operations. But these technologies are years away from being deployed at the scale needed to offset the AI-driven surge in demand.
The Industry-Wide Challenge
The energy question is becoming one of the defining challenges of the AI era. The technology promises to accelerate scientific discovery, transform industries, and boost economic productivity — but it also threatens to strain the world’s energy infrastructure in ways that are only beginning to be understood. Google’s 37% jump is a data point in a much larger story about whether the AI industry can grow sustainably, or whether its environmental costs will become a constraint on its ambitions.
For now, the trend line is clear: AI is getting bigger, models are getting hungrier, and the power grids that feed them are running faster just to keep up. Google’s environmental report is a reminder that the AI revolution has a physical footprint — and it is growing fast.