Kenya Warns Microsoft That $1 Billion AI Data Center Could Consume Half the Nation’s Electricity
May 17, 2026
Kenya Warns Microsoft That $1 Billion AI Data Center Could Consume Half the Nation’s Electricity
The rapid expansion of artificial intelligence infrastructure is colliding with real-world energy constraints, and Kenya has become the latest flashpoint. Government officials have warned that Microsoft’s proposed $1 billion AI data center, developed in partnership with Abu Dhabi-based AI firm G42, could require so much power that it might necessitate "switching off half the country" to keep the facility operational. The warning underscores a growing global tension between the energy demands of AI and the capacity of national power grids.
The project, originally announced as a major Azure cloud and AI region for East Africa, was planned to run on geothermal energy from Kenya’s Rift Valley. Initial designs targeted a capacity of roughly 100 megawatts, with long-term ambitions scaling up to 1 gigawatt. However, Kenya’s peak electricity demand has already reached approximately 2,444 megawatts earlier this year, according to a Bloomberg report. A fully scaled 1 GW AI facility would therefore consume a substantial portion of the nation’s available power infrastructure, raising serious concerns about grid stability and equitable energy distribution.
Negotiations between Microsoft, G42, and Kenyan authorities have reportedly stalled over power guarantees and infrastructure commitments, though officials maintain that the project has not been canceled outright. The situation highlights a broader challenge facing the global AI industry: as models grow larger and chips become faster, the electricity required to run them is escalating far faster than many national grids can accommodate. Data centers already consume a significant share of electricity in major markets, and Kenya’s predicament may serve as a preview of similar conflicts elsewhere.
The uncomfortable reality is that AI’s massive energy demands are becoming harder to ignore worldwide. Kenya’s warning, while stark, reflects a question that many countries will soon have to confront: powering the AI boom is increasingly an infrastructure problem as much as a technological one. Without significant investment in grid capacity and renewable energy sources, the race to build ever-larger AI facilities may face mounting resistance from governments trying to balance innovation with the basic needs of their citizens.
Source: digitaltrends
