Electricity costs, environmental effects top of mind for British Columbians opposed to AI data centres

A new poll from Leger documents British Columbians' concerns over artificial intelligence data centres in their province.

British Columbians are evenly split on whether building AI data centres in Canada is a "good thing" or a "bad thing," 42% vs 44% respectively. Yet national figures do not tell the whole story.

While British Columbian might support data centres elsewhere in Canada, a strong plurality of 49% oppose the development of data centres in BC itself, while 37% support.

When asked specifically about the concerns over AI data centres, the reasons for local opposition becomes much clearer. A full 88% of British Columbians say they are concerned data centres will increase their home electricity costs, while 85% cite concern over environmental impacts such as water use and greenhouse gas emissions.  

In May, Telus announced two new data centres to be built in Vancouver.

WATCH: New AI data centres to be built in BC as feds push for ‘digital sovereignty’

“We can’t run a sovereign AI strategy on someone else’s servers, governed by someone else’s rules, and in someone else’s jurisdiction.”

Sitka MediaJarryd Jäger

The city has also witnessed street protests against data centre development.

WATCH: Protest against AI data centres in Vancouver draws hundreds

Many expressed aversion to artificial intelligence itself, arguing that the costs of the technology outweigh the benefits.

Sitka MediaJarryd Jäger

BC Green Party Leader Emily Lowan has specifically called for a moratorium on data centre development.

WATCH: BC Greens call for moratorium on AI data centres until ‘regulatory safeguards’ in place

The move comes on the heels of an announcement that numerous facilities will be built across BC in the near future.

Sitka MediaJarryd Jäger

Methodology: Leger’s study is based on an online survey of 1,505 Canadian residents aged 18 and older, conducted from July 10 to 13, 2026, using Leger’s LEO online panel. Results were weighted according to age, gender, mother tongue, region, education, and presence of children in the household to ensure a representative sample of the Canadian population. A margin of error cannot be associated with a non-probability sample, but for comparison, a probability sample of this size would have a margin of error of ±2.53%, 19 times out of 20.