The short version
- The global number of data centers has surged from approximately 8,000 in 2021 to over 12,000 today, with the United States hosting nearly half of all operational sites.
- Opposition to new facilities is intensifying due to concerns about electricity consumption, water usage, and air pollution, leading sixteen US states to propose legislative bans or delays.
- While industry leaders argue these facilities are vital for national security and economic growth, critics point to the environmental costs and limited local job benefits of certain facility types.
The rapid expansion of data centers has evolved from a largely invisible background process into a prominent political flashpoint as the United States approaches its midterm elections in November. These massive facilities, which house the servers and advanced computer chips necessary for artificial intelligence services and general internet activity, are now at the center of a heated debate involving environmental impact, resource allocation, and economic policy. The issue has drawn sharp divisions between technology advocates and local communities, with both Republican and Democratic lawmakers expressing skepticism toward unchecked development.
The scale of this infrastructure boom is substantial. Global tracking data indicates that there are currently around 12,250 operational or pre-operational data centers worldwide, a significant increase from the roughly 8,000 sites recorded in 2021. The United States remains the dominant hub for these facilities, containing approximately 4,767 locations. This concentration is expected to widen the gap between American capacity and that of other regions, including Europe and China, raising concerns among some analysts that other nations may miss out on the economic advantages associated with the AI revolution.
Historically, data centers were situated near urban areas to facilitate access to high-voltage power lines and water resources required for cooling systems. However, as facilities grow in size and complexity, developers are increasingly turning to rural locations. This shift has brought industrial-scale computing infrastructure into closer proximity with residential communities, sparking friction over land use and local environmental conditions. The physical footprint of these sites varies dramatically, ranging from small modular units to sprawling campuses that cover millions of square feet.
Resource consumption is a primary driver of public concern. Data centers worldwide are estimated to consume more electricity than the entire nation of France, a demand that continues to rise as artificial intelligence models become more complex. This intense energy appetite places significant strain on local power grids and contributes to increased carbon emissions. Additionally, the cooling processes required to maintain server temperatures can lead to substantial water usage, raising fears among experts about potential shortages in arid regions or areas with limited water infrastructure.
Specific projects have highlighted these tensions. For example, the construction of large-scale AI campuses involves housing hundreds of thousands of advanced chips and utilizing significant power sources, including methane gas turbines in some cases. Activists have argued that such measures could degrade local air quality, although regulatory bodies maintain that emissions will remain within established standards. The noise generated by cooling systems and the visual impact of warehouse-like structures have also become points of contention for residents living near new developments.
The political response to these concerns is becoming increasingly organized. Sixteen US states have introduced legislation aimed at banning or temporarily halting the construction of new data centers. This legislative activity suggests that the issue could play a decisive role in upcoming elections, pitting local opposition against federal and industry support for technological advancement. President Donald Trump has characterized the resistance to data center expansion as a conspiracy, further polarizing the debate between executive branch priorities and grassroots environmental advocacy.
Proponents of the industry emphasize the strategic importance of these facilities for national security and the broader economy. They argue that data centers attract significant investment, generate tax revenue, and create employment opportunities in construction and maintenance sectors. However, research from the Brookings Center indicates that the economic benefits are not uniform across all facility types. Hyperscale centers operated by major technology firms tend to provide more substantial local job growth compared to colocation facilities, which lease space to remote tenants without maintaining a significant on-site workforce.
Despite the growing backlash, investment in AI infrastructure continues to accelerate. The technology sector is projected to spend approximately $765 billion on data center development in 2026, with figures expected to exceed one trillion dollars in the following year. In response to environmental pressures, some industry players are exploring unconventional solutions, such as locating facilities in space or on the moon to utilize abundant solar energy. While these ideas remain speculative, they underscore the urgency with which the tech industry is seeking to reconcile its expanding footprint with sustainability concerns.
As the midterm elections approach, the debate over data centers reflects a broader struggle to balance technological innovation with environmental stewardship and community well-being. The outcome of this political contest may determine the regulatory landscape for AI infrastructure in the coming years, influencing where and how these critical facilities are built. With global demand for computing power showing no signs of slowing, the tension between rapid deployment and responsible management is likely to persist.
Sources behind this briefing
Go to the original reporting
- BBC News↗Data centre debate heats up ahead of US midterms - here's what you need to know
