Opposition to AI data centres is surging across Canada as the number of proposed facilities climbs to 96, according to a study from York University. While these centres promise technological growth, residents and local governments are increasingly pushing back, citing concerns over massive electricity, water, and land consumption. With only five hyperscale facilities currently operating in the country, the scale of upcoming development has triggered widespread environmental and social anxiety.
Why are communities protesting AI data centre projects?
Public opposition is rooted in the physical footprint of these facilities. In Vancouver, residents marched in late May to protest two planned centres, highlighting concerns that the infrastructure demands too much water and energy during times of regional restriction. Similar protests occurred in April at the Saskatchewan legislature, while in Olds, Alberta, a local group organized specifically to block a proposed data centre development.
The intensity of this pushback is already yielding results. In Hamilton, Ontario, a planning tribunal denied a developer’s application to split land for a “hyperscale and enterprise” data centre after hundreds of residents raised fears regarding local health, pollution, and the environmental impact of AI. In Manitoba, Premier Wab Kinew confirmed a large proposed data centre south of Winnipeg would not proceed, citing concerns for both the environment and the rural community. An online petition against that project had garnered over 13,500 signatures.
A single query run through a generative AI chatbot like ChatGPT requires nearly 10 times more electricity than a standard Google search, according to research cited in the York University study.
How much power and water do these facilities actually consume?
The sheer scale of resource consumption is a primary driver of the debate. According to a June 3 report from the United Nations University Institute for Water, Environment and Health, data centres consumed an estimated 448 terawatt-hours of electricity globally in 2025. If that consumption were a nation, it would rank as the 11th largest in the world.
The York University study estimates that a typical 100-megawatt data centre requires between 438,000 and 700,800 megawatt-hours annually—roughly equivalent to the electricity usage of 40,000 to 64,000 Canadian households. Beyond power, the UN report found that global data centres used 4.5 trillion litres of water in 2025 for cooling systems, a volume that could meet the needs of more than 600 million people in Sub-Saharan Africa.
Can technology mitigate the environmental impact?
Industry leaders argue that modern engineering can significantly reduce the strain on local utilities. Todd Coleman, CEO of the Montreal-based company eStruxture, stated that his facilities use closed-loop glycol cooling systems, meaning they do not consume water beyond basic facility needs like sinks and toilets. Similarly, Matt Milton, president of Microsoft Canada, noted that the company is deploying new techniques to lower water usage, and a Utah-based project backed by investor Kevin O’Leary is utilizing water-recycling closed-loop technology.
Is the backlash about more than just the environment?
Blayne Haggart, a political science professor at Brock University, suggests that data centres have become physical targets for broader public anxiety regarding the rapid rise of AI. This frustration is compounded by recent job losses in the tech sector; companies like Meta, Block, Amazon, and Microsoft have all reduced their workforces while shifting focus toward AI.
Public sentiment appears to support stricter oversight. A recent Angus Reid poll found that 68 per cent of respondents would oppose a large AI data centre within a few blocks of their home. Furthermore, 52 per cent of those polled believe AI data centres are a negative for job creation, compared to just 16 per cent who view them as a benefit. While the federal government’s new AI strategy aims to create 250,000 jobs through AI adoption, these figures suggest a significant gap between official policy goals and public perception.
Frequently Asked Questions
- Are data centres regulated in Canada? The federal government has recently unveiled an AI strategy that includes plans for large-scale data centres and AI literacy training, though opposition parties have questioned the job-creation projections.
- Do all data centres use large amounts of water? Not necessarily. Industry representatives state that modern facilities are increasingly moving toward closed-loop cooling systems that significantly reduce or eliminate external water consumption.
- Why do people protest data centres? Concerns typically focus on the high electricity and water consumption, potential local pollution, and broader anxiety regarding AI’s impact on employment and the environment.
What are your thoughts on the growth of AI infrastructure in your community? Share your perspective in the comments below or subscribe to our newsletter for more updates on the intersection of technology and local policy.
