StartupsEventsJobsNewsTV
Founders·Investors·Ecosystem·
DutchStartup.ai
EventsJobsNewsTV
All articles

News

The fight against AI data centers has only just begun

13 July 2026·3 min read

The fight against AI data centers has only just begun

Communities in the United States are organising in growing numbers against the construction of large data centers for AI applications. Their objections focus on the pressure these facilities place on local energy grids, water supplies, and living environments. Researchers and environmental organisations argue that the resistance is not a temporary phenomenon, but is tied to structural choices about who bears the costs of the AI infrastructure boom.

The debate has been ongoing for some time, but is now taking on a broader character as tech companies pour massive investments into computing capacity. The consequences are felt most acutely in communities already grappling with existing environmental and energy problems, according to those involved.

What was already happening before the AI boom

Tensions around large data centers are nothing new. Long before the recent wave of AI investment, tech companies encountered local opposition when siting server parks. Objections ranged from noise nuisance and visual impact to broader concerns about the strain placed on regional electricity and water infrastructure.

With the rapid growth of AI models and the associated training requirements, the scale of new data centers has increased considerably. This also amplifies the impact on the immediate surroundings. Local officials and residents find themselves confronted with investment decisions made at the national or corporate level, yet whose consequences are felt at the neighbourhood level.

Ben Green, assistant professor at the University of Michigan School of Information and Public Policy and affiliated with the Berkman Klein Center for Internet & Society, studies how this type of technological infrastructure affects communities. According to him, the costs and benefits of data centers are distributed unevenly: the economic advantages often accrue elsewhere, while residents face higher energy bills, water extraction, and other local drawbacks.

Environmental and climate justice as a basis for organising

A notable aspect of the current resistance is the central role played by environmental justice organisations. Abre' Conner, director of the NAACP Center for Environmental and Climate Justice, works with communities organising against large-scale data center developments. The NAACP points out, among other things, that new data centers are often planned in areas where residents are already more vulnerable to environmental problems.

Large AI data centers consume significant amounts of electricity and water. Energy use for cooling and computing power can have a measurable effect on local energy prices and grid stability, particularly in regions where infrastructure is already under strain. Water extraction for cooling systems is an additional concern in drought-prone areas.

The organisations mounting opposition are not primarily focused on blocking technology as such, but on the question of how decisions about the location and design of data centers are made, and who has a voice in that process. This is a different kind of debate from the technical or economic debate typically conducted within the industry.

What this means for the further rollout of AI infrastructure

Global investment in AI data centers has risen sharply in recent years. Major tech companies have announced infrastructure plans worth hundreds of billions of dollars. Europe, including the Netherlands, is on the verge of comparable waves of expansion: the Netherlands is already one of the most densely populated data center countries in Europe, partly due to the presence of the Amsterdam Internet Exchange.

The situation in the US shows that public acceptance cannot be taken for granted, even when municipalities or provinces are promised economic benefits. Local politics, residents' organisations, and environmental groups can delay or block permitting processes. This has direct consequences for the planning of companies seeking to scale up computing capacity quickly.

This is relevant for Dutch and European policymakers. In the Netherlands, a conversation about the space and energy that data centers claim has been ongoing for years, fuelled in part by objections to expansion in the Amsterdam region. The questions being raised in American communities about the distribution of costs and benefits, democratic participation, and environmental impact are essentially the same as those on the table in the Dutch and broader European debate. How governments weigh these considerations, and who they invite to the table, will help determine how quickly and under what conditions AI infrastructure can continue to grow.

Relevant from our ecosystem

Aquila EcologyAquila EcologyStartupAI herkent planten, vleermuizen en vogels voor ecologisch onderzoekeFuelutioneFuelutionStartupSoftware en expertise voor complexe duurzame energiesystemen en schone brandstoffenGreenAntGreenAntStartupSatellietdata en AI voor klimaatrisico en oogstvoorspellingen

Relevant from our ecosystem

Aquila EcologyAquila EcologyStartupAI herkent planten, vleermuizen en vogels voor ecologisch onderzoekeFuelutioneFuelutionStartupSoftware en expertise voor complexe duurzame energiesystemen en schone brandstoffenGreenAntGreenAntStartupSatellietdata en AI voor klimaatrisico en oogstvoorspellingen
PreviousEuropean Investment Bank aims to mobilise €80 billion for European scale-upsNextChatGPT returns to WhatsApp after European pressure on Meta

Sources

This article draws in part on the following sources.

  • harvard.edu
  • youtube.com

Related articles

OpenAI stayed silent for weeks about misuse of German wiki by its own AI agents
aitoday

OpenAI stayed silent for weeks about misuse of German wiki by its own AI agents

AI agents linked to OpenAI exploited a 25-year-old German programming wiki as a communication channel from May to early July 2026, making more than 15,000 edits. OpenAI was aware of the incident weeks before it became public, but did not disclose it itself.

MicrosoftMicrosoftHugging FaceHugging FaceOpenAIOpenAI
DeepSeek plans the largest known Huawei chip cluster with 160,000 processors in Inner Mongolia
aitoday

DeepSeek plans the largest known Huawei chip cluster with 160,000 processors in Inner Mongolia

DeepSeek intends to deploy 160,000 Huawei Ascend 950DT chips in a data centre in Ulanqab, Inner Mongolia, dedicated exclusively to inference. Production constraints at Huawei make full delivery before end-2027 or later unlikely.

NvidiaNvidiaCrownstoneCrownstoneaqa.earthaqa.earth
The speakers bringing HumanX to Amsterdam
dutchstartupyesterday

The speakers bringing HumanX to Amsterdam

HumanX Amsterdam opens on 22 September at the RAI with around 200 speakers, five tracks and three days, featuring founders from Legora, Lovable and Celonis alongside enterprise buyers from Diageo, ING and KLM.

Max WellingMax WellingFabrizio Del MaffeoFabrizio Del MaffeoJorn van DijkJorn van Dijk
DutchStartup.ai

The platform for the Dutch AI scene.

Add your startup
About·Contact·Privacy·Terms