The Data Center Debate | News Analysis of Canada's Energy Misinformation
In today’s episode, The Sanity Project delivers a sharp news breakdown focused on critical thinking—taking listeners beyond headlines to dissect the narratives shaping Canadian current events. Hosts Rachel Bennett and Michael Reeves cut through the noise about EVs and AI data centers, revealing what’s actually straining Canada’s power grid and why these stories matter for everyone interested in responsible journalism and rigorous public discourse.
To subscribe to our free weekly newsletter, go to https://thesanity.org/subscribe
Cutting Through Outrage Culture: A New Kind of Podcast Sharp Political Analysis for Canadian NewsThe Sanity Project is your destination for incisive news analysis and fearless political commentary on the latest Canadian news. We dive into politics with a firm commitment to critical thinking—challenging narratives and exposing media misinformation. Whether you lean liberal or are simply seeking smarter conversation, you’ll find our approach refreshingly unswayed by partisan noise.
Combatting Misinformation in Current EventsTired of outrage culture distorting democratic debate? We unravel stories swirling through the daily news—especially those most politicized in Canada. From dissecting grid anxiety myths to showing how progressive politics interacts with technology and energy, our news commentary goes deeper, always keeping you ahead of the narrative curve.
Why Smart News Commentary MattersCanadian politics is complex. We make sense of fast-moving current events for listeners who care about truth and transparency in a democratic society. With every episode, The Sanity Project delivers the political analysis and nuanced insights you won’t find in most Canadian news feeds—all built on facts, not fear.
To never miss an episode, subscribe to this podcast wherever you are listening.
Everyone's been told electric vehicles are about to crash the grid, overload the system,
and drive up your power bill.
But what if that entire narrative is wrong?
And what if the real threat to Canada's power system isn't sitting in your driveway,
but inside a windowless warehouse?
Hi, I'm Beau Kaufman, and this is The Sanity Project.
Today, we're diving into one of the most persistent and politically convenient narratives
in Canada right now, that electric vehicles are going to overwhelm our power grid and
trigger a wave of instability across the country.
It's a claim you've heard everywhere, from media headlines to political debates to social
media commentary.
But when you actually look at the data and start asking who benefits from that narrative,
the story begins to shift in some very uncomfortable ways.
So we've taken this white paper by Our Sanity Project and turned it into a full, deep-dive
debate.
Setting the Debate — Who Benefits?
Senior anchor Rachel Bennett is here to walk us through what the data actually says, while
investigative reporter Michael Reeves has been digging into the contradictions, policy
decisions, and priorities that aren't being discussed.
Because this isn't just about energy.
It's about misinformation, it's about incentives, and it's about what's really putting pressure
on Canada's grid right now.
Let's listen in.
So right now, every day people are being warned that just plugging an electric vehicle into
their garage is going to somehow trigger nationwide blackouts.
It's a narrative you hear pretty much everywhere.
Exactly.
But at the exact same time, utilities are quietly spending billions of dollars to refurbish
aging coal plants.
And they're doing it to power this massive new invisible energy glutton.
Right.
A sector that requires basically the power of 15 small towns just to run one single building.
So welcome to today's Deep Dive.
Our mission today is to untangle one of the most, well, persistent narratives dominating
the news right now.
The claim that widespread EV adoption is just going to completely overwhelm and collapse
our power grids.
Right.
Because whether you're, you know, actively shopping for a new car, or you're just staring
at your monthly utility bill wondering what the future holds, you're being sold this story
of imminent infrastructure doom.
Yeah, total doom and gloom.
So today, we're looking at the hard data to show why that story is just mathematically
flawed.
And more importantly, what is actually putting all this strain on the grid?
Yeah.
And to get to the bottom of this, we're relying on a really robust stack of sources today.
What are we looking at?
So we've got a May 2026 white paper from the Sanity Project.
That one's called The Grid Myth.
We also have Natural Resources Canada's Powering Canada's Future Strategy, a June 2024 congressional
report from the U.S. Department of Energy, and this really fascinating policy analysis
from Policy Options.
Right.
So we're looking at that whole situation unfolding in Saskatchewan.
Exactly.
And this stack is great because it lets us examine the issue from, like, the microscopic
engineering level all the way up to macroeconomic policy.
Okay, let's just dive right into the primary myth we hear everywhere.
Social media, the news, legislative debates.
It's this idea of the 5 p.m. shock.
Ah, right.
The 5 p.m. shock.
Yeah, like everybody drives home from work, everyone plugs in their car at the exact same
moment, and the grid just simply melts down, right, sending rolling blackouts across the
whole country.
Which, I mean, it plays perfectly into our basic intuition about electricity, right?
Totally.
I mean, we've all tripped a circuit breaker in our house by, I don't know, running a hairdryer
and a space heater at the same time.
Yep, done that.
So scaling that up to millions of vehicles just feels instinctively terrifying.
It does.
But looking at the actual data from the International Energy Agency fundamentally changes that picture.
IEA Data: EVs’ Small Share of Demand
Okay, what does the IEA say?
Well, globally, electric vehicles currently represent a mere 0.5% of total electricity
demand.
Wait, half of 1%?
That's it?
That's it.
Today, it's just 0.5.
And the IEA projects that even under their most aggressive, you know, high-adoption scenarios,
looking all the way out to 2035 EVs, we'll only reach about 6 to 8% of total global electricity
demand.
Okay, but I mean, I don't really understand how a 12-year timeline solves the localized
problem.
How do you mean?
Because 8% of the entire grid still sounds incredibly massive.
If I add an 8% electrical load to my house, say I build a whole new addition, and I just
expect my existing old wiring to handle it, I'm going to start a fire.
So even if I have a 12-year warning, the ultimate capacity of the wire coming into my neighborhood
hasn't magically changed.
I see what you're saying.
Are utility companies going to have to like dig up every single street in the country
to lay thicker cables?
No, and that's the key here.
Why Utilities Don’t Need to ’Dig Up’ Every Street
Utilities don't need to dig up every street simultaneously because grid infrastructure
isn't this static, permanent installation.
Right, it's always being worked on.
Exactly.
It's constantly degrading, being maintained, being replaced.
Like transformers, those big metal cylinders you see up on telephone poles, they have natural
lifespans.
Oh, so they eventually just break down anyway.
Yeah, and grid operators work on these multi-decade planning cycles.
Transformer Lifecycles & Planning Cycles
So when they know an 8% demand increase is coming over this 12-year slope, they use predictive
modeling.
To see who's buying the cars.
Exactly.
Which neighborhoods are adopting EVs the fastest?
And then, when an old transformer in that specific neighborhood is scheduled for its
routine replacement anyway, they just swap it out for a larger capacity model.
Oh, I get it.
So they're just kind of riding the natural replacement cycle of the equipment?
Exactly.
They're absorbing the upgrade into standard operational planning.
Yeah.
I mean, think about the massive adoption of central air conditioning back in the 60s and 70s.
Oh, wow.
Yeah.
That must have been a huge spike.
Huge.
From zero home cooling to millions of these massive compressors pulling huge amounts of
power on the hottest days of the year.
Or even the explosion of home computing, right?
And the grid didn't collapse.
Exactly.
Because utility planners anticipate demand growth.
They build it into their capital expenditure models.
So this premise that grid planners somehow just forgot to model for EV adoption, it's
a rhetorical device used to generate anxiety.
It's not an actual engineering reality.
Not at all.
That honestly reframes the whole capacity argument for me.
It's, you know, it's not a surprise party.
It's an RSVP'd event that the host has a full decade to prepare for.
That's a great way to put it.
Yeah.
But, OK, that still is the timing issue, right?
Even with upgraded transformers, if everyone pulls electricity at the absolute peak hour
of 5 p.m., that surge really must outpace whatever generation is happening at the power
plant.
Well, the 2024 U.S. Department of Energy congressional report actually looked directly at that.
DOE Findings: When People Actually Charge EVs
They looked at the actual charging behavior of EV owners moving past all these theoretical
worst case scenarios.
And what did the actual data show?
It shows the vast majority of charging does not happen at 5 p.m.
Really?
Yeah.
It happens overnight, primarily between 10 p.m. and 6 a.m.
Oh, so people just plug in right before bed and the car just sits there in the garage
for eight hours.
Exactly.
And crucially, from an engineering standpoint, this overnight window is exactly when grid
demand is at its absolute lowest.
Because everyone's asleep.
Right.
Everyone's asleep.
Businesses are closed.
Industrial demand drops way down.
But our baseload generation cannot easily just turn off.
What do you mean by baseload generation?
So when we talk about baseload, we're talking about massive thermal power plants, things
like nuclear facilities or huge coal plants.
They're basically the freight trains of the energy world.
Oh, so they take a while to get going.
Yeah, it takes massive amounts of fuel and time to get a giant steam turbine spinning
up to speed.
Yeah.
You cannot just hit the brakes at midnight because people went to sleep.
They really prefer to keep running at a steady speed 24-7.
Wow.
So the plants are just churning out gigawatts of power into a grid where, frankly, nobody
is home to turn on a light switch.
Exactly.
It creates this massive surplus of electricity.
In some wholesale markets, power prices actually go negative at night.
Wait, negative?
They pay people.
Literally, yeah.
Because plants would rather pay someone to take the electricity than incur the mechanical
damage and the huge cost of shutting down and then restarting their turbines the next
morning.
That is wild.
And add to that the fact that wind generation naturally peaks at night when the thermal
currents shift.
So you have a system that is just flooded with cheap, unused power.
OK, so electric vehicles aren't a burden on the system at all.
They're basically acting as a giant sponge for energy that would otherwise just be totally
wasted.
Precisely.
The DOE report specifically uses the word symbiotic to describe this relationship.
EVs as an Energy ’Sponge’ and Smart Chargers
Symbiotic.
Yeah.
EDs act as this flexible distributed load.
So instead of stressing the system, they actually stabilize it by providing a place
for all that excess base load and wind generation to go.
It's essentially like programming your dishwasher to run at 2 a.m.
Yes, exactly.
You get the exact same result, clean dishes or a fully charged car, but you are utilizing
the plumbing when absolutely no one else in the house is competing for the water pressure.
Right.
So I got to say, speaking from experience, if you run a super loud dishwasher at 2 a.m.,
the grid might be totally fine, but your spouse is definitely going to stage a blackout of
their own.
Probably not a great idea.
But the beauty of EV charging is that it's completely silent and it doesn't even rely
on human behavior to manage it.
Right.
Because of smart chargers.
Exactly.
Smart chargers are standard in almost all current home installations now.
You plug the car in at 5 p.m. when you get home from work, but the car and utility communicate.
The charger is literally programmed to wait until, say, 11 p.m. to actually pull the power.
So the simultaneous 5 p.m. charging spike is just a solvable rate design problem.
Completely solvable.
It's not a fundamental crisis.
You know, think about your own daily routines for a second.
Our listeners, I mean.
How much of your energy, like your laundry, heating your water, running the dishwasher,
how much of that could actually be shifted to off-peak hours if it just happened automatically
in the background?
It's an interesting question.
Yeah.
Let us know what you think, because we honestly love hearing how you are all navigating these
shifts in your own homes.
Smart Charger flexibility just seems like the ultimate hack to keep infrastructure costs down.
Oh, it is.
It's literally the foundation of modern grid management.
So smart chargers solve the timing issue by waiting until 2 a.m., but you know, that still
leaves a lingering guilt for a lot of drivers.
It's emissions guilt.
Exactly.
Because if you're plugging in at night and you're acting as a sponge for baseload power,
the argument is that you are essentially just fueling your clean car with a dirty coal plant.
Right.
The Sanity Project white paper calls this the coal car in disguise narrative.
Yeah, it's a very common talking point.
But addressing that requires looking at the specific context of where you are.
And that's heavily detailed in Natural Resources Canada's Powering Canada's Future strategy.
Canada’s Clean Grid: 82% Non‑Emitting
Okay, what does the Canadian data show?
Well, the reality of Canada's electrical grid completely shatters the coal in disguise argument.
Their data confirms that 82% of Canada's electricity generation comes from non-emitting sources.
82% non-emitting.
That is a staggering number.
It really is.
Breaking down the generation mix, you've got 60% coming from hydro, 14% from nuclear, and
about 8% from wind, solar, and other renewables.
Yeah, it makes Canada one of the cleanest electricity systems of any industrialized nation.
So if you are driving an EV in Canada, statistically speaking, you are definitely not driving a
coal car.
You're overwhelmingly driving a hydro car.
Basically, yeah.
It's just, you know, water spinning turbines inside a dam, sending power down a line to
charge your battery.
Exactly.
And in provinces like Quebec, Manitoba, and British Columbia, where the grid is almost
entirely hydroelectric, the life cycle emissions of an EV drop to numbers that approach absolute zero.
That's incredible.
But we do have to look at the whole map, right?
Because Canada is a massive country and the grid isn't identical from coast to coast.
No, it's not.
And that's a really important point to make.
The coal argument absolutely does have geographic merit in specific locations.
Like where?
Well, in provinces like Saskatchewan and Alberta, the grid mixes remain highly carbon-intensive.
They rely heavily on natural gas and, historically, coal.
So an EV charged in Regina today objectively carries a different carbon footprint than
an EV charged in Montreal.
So the power source really matters locally.
It does.
But the policy options analysis points out the danger in how this data is deployed politically.
What do you mean?
Well, they note that taking the localized, fossil-heavy grid profile of a specific province
like Saskatchewan, and then applying it as a blanket argument against national EV adoption,
that's a highly selective use of statistics.
Oh, I see.
It takes a regional generation challenge and frames it as a national liability.
When a country's grid is 82% clean overall, arguing that EVs run on coal is mathematically
disingenuous on a macro level, we're just reporting what the analysis found here.
Staying neutral, but the math is what it is.
Right, right.
Let's follow the logic here, then.
So the math proves EVs represent a slow, manageable 12-year growth curve.
And the physics show they charge mostly off-peak and actually help stabilize baseload generation.
And the generation data proves they're overwhelmingly running on clean power.
So if EVs are highly predictable and grid-friendly, what is actually causing all this grid anxiety
we keep hearing about?
Well, this is where it gets really interesting.
The International Energy Agency data points to a very different, completely unprecedented
stressor on our electrical infrastructure right now.
Real Grid Stressor: AI Data Centers
And what's that?
It's artificial intelligence data centers.
AI data centers?
Yes, yes.
In a single year, electricity consumption from AI-focused data centers surged a staggering
50%.
A 50% jump in 12 months.
Just 12 months.
That's insane.
Why is the power draw so violent compared to just, you know, normal internet use?
Because the computing is completely different.
Like, a traditional internet search is essentially a retrieval task.
It requires very little energy.
You search for a recipe, it grabs a link.
But generative AI fundamentally changes the computing mechanism.
An AI query requires roughly 10 times the electricity of a typical search engine query.
Wait, 10 times?
For one search?
Yeah, because the servers are calculating probabilities across billions of parameters
in real time.
Oh, wow.
It's like the difference between asking a librarian to fetch a book that already exists
on a shelf versus asking that same librarian to write a brand new original book for you
on the spot before you leave the desk.
That's a perfect analogy.
Generating that new information requires massive computing power.
And the GPU clusters that do that math, they generate immense, intense heat.
Which means they need cooling.
Exactly.
It requires massive industrial air conditioning and water cooling systems just to keep the
building from melting down.
That compounding effect is why the power draw is 10 times higher.
That is unbelievable.
And overall, data center demand is projected to double by the end of 2026.
This is a demand surge with virtually no precedent in the modern grid era.
And unlike EVs, AI data centers are completely inflexible.
They require immense, always-on-base load power.
They run at maximum capacity 24 hours a day, seven days a week.
You cannot ask an AI server farm to just wait until 2am to process data.
Because everyone's querying AI all day long.
So this brings us back to that glaring contradiction unfolding in Saskatchewan that we mentioned
earlier.
Yes, the policy auctions piece.
Right.
Saskatchewan is a province where the leadership has been actively warning everyday citizens
that the grid cannot handle electric vehicles.
Yet the analysis details a completely different reality happening on the industrial side.
Yeah, it's a documented policy contradiction.
Saskatchewan Case: Bell Canada AI Campus Announced
And it really centers around an announcement from Bell Canada in March 2026.
Oh, they announced.
They are building a $1.7 billion AI data center campus near Regina.
And this single facility will draw 300 megawatts of constant power from the provincial grid.
300 megawatts?
I mean, that's difficult to even conceptualize.
Didn't the DOE report note that a typical small town might use about 20 megawatts?
Yeah, about 20.
So this single AI facility is pulling the power equivalent of 15 entire small towns
all day, every day, without stopping.
Exactly.
And power availability was actually cited as the primary reason the tech company chose
Saskatchewan in the first place.
Interesting.
The government is actively committing major crown infrastructure to support this 300 megawatt campus.
Like what kind of infrastructure?
Well, SaskPower is building the dedicated transmission lines.
SaskTel is laying the necessary fiber optics.
And SaskEnergy is constructing a brand new high-pressure gas pipeline straight to the site.
The messaging is completely fractured there.
I mean, the public is told the grid is far too fragile to handle someone plugging a Chevy
Bolt into a 220-volt outlet in their own garage, right?
Right.
And when they say that, the utility is actively trenching new high-pressure gas pipelines
and stringing heavy transmission lines to feed this invisible energy glutton.
And the situation becomes even more complex when we look at the generation side of it.
How so?
Well, to guarantee that 300 megawatts of 24-7 baseload tower, SaskPower is relying heavily
on fossil fuels.
In their recent filing for a rate increase, SaskPower disclosed a $2.6 billion coal refurbishment
program.
Wait, $2.6 billion?
Yes.
And the prime minister openly credited the decision to refurbish these aging coal plants
with making the data center deal possible.
So a multi-billion dollar effort to extend the life of coal plants specifically to power
AI computing.
Who is paying for a huge capital expenditure like that?
Well, SaskPower is seeking rate increases of nearly 4% in both 2026 and 2027 to help
cover these capital investments.
Let me make sure I have this straight.
Who Pays? Rate Hikes and Corporate Subsidies
Very residential rate payers, like families running their laundry, small businesses trying
to keep their lights on, they are facing utility rate hikes to upgrade coal plants.
So a massive telecommunications giant can run an AI server farm.
According to the filings, yes.
And those same everyday citizens are simultaneously being told that their electric cars are the
real threat to the grid.
Exactly.
And, you know, the analysts at PolicyOptions frame this purely through the lens of grid
economics.
It's like surge pricing on an Uber.
OK, sure.
When supply and supply is scarce, the price goes up to manage the system.
But Saskatchewan currently lacks a standing framework to appropriately price scarce, reliable
baseload power.
So they just sell it cheap.
Right.
When a utility sells its most valuable 24-7 reliable electricity at a standard posted
rate to a massive hyperscale tech company, the utility is essentially subsidizing that
corporation.
Right.
And the cost of the necessary grid upgrades and the coal refurbishments, they're then
passed on to the rest of the ratepayer base.
They're subsidizing the tech giant and then handing the bill for the infrastructure to
the person trying to charge their car.
It really all comes down to what the system prioritizes.
It does.
And the Sanity Project White Paper concludes that this isn't an engineering argument at
all.
It's a priority argument.
Right.
Their actions and infrastructure commitments reveal that current policy priorities heavily
favor these hyperscale AI campuses over ordinary citizens attempting to electrify their transport.
So when we assemble all the pieces of this puzzle, a very clear picture emerges, doesn't
it?
Priority, Not Engineering — Sanity Project Conclusion
We have electric vehicles sitting at just 0.5% of global demand, acting as this symbiotic
sponge for off-peak power on a Canadian grid that is 82% clean.
And then on the other side, we have AI data centers driving a 50% surge in power consumption,
requiring billions of dollars in coal refurbishments and residential rate hikes just to keep their
servers cooled down.
So the verdict from the engineering and economic data is pretty definitive.
The narrative that electric vehicles will break the power grid is just a political distraction.
The math proves that EVs are highly manageable, they're predictable, and they're uniquely
grid friendly.
So the true unprecedented stress on our electrical infrastructure, the demand that is requiring
immediate massive capital investments and the extension of fossil fuel lifespans, that
is coming from AI data centers.
Absolutely.
The real grid stress is coming from servers, not drivers.
We've been pointed toward our own driveways to find the culprit when we should be looking
at the server farms processing our daily internet queries.
Exactly.
Final Verdict — Servers, Not Drivers
You know, if you want to read more data-driven deep dives just like this one, really shedding
light on the actual mechanisms and math behind the headlines, please hit subscribe on this
show.
And make sure you check out blog.thesanity.org.
They do incredible work untangling these complex issues so you can actually understand the
physics and the economics at play.
Yeah, looking past the rhetoric to follow the megawatts.
It's just essential right now.
And I want to leave you with one final broader consideration.
Oh, lay it on us.
Well, we've spent this time examining how artificial intelligence is drawing immense
unprecedented power from our grids.
But we are also increasingly told by the tech industry that AI is the critical tool we need
to solve our most complex logistical problems.
Oh, right.
Like how to build a smarter, more efficient electrical grid.
We're supposed to use AI to optimize the grid itself to figure out where the power needs
to go.
Exactly.
And yet, AI tools require 10 times the power of normal computing just to generate an answer.
Every single time we use AI to try and optimize our lives or optimize our power grids.
Are we inadvertently accelerating the exact energy crunch we're trying to avoid?
Wow.
We started this deep dive talking about the grid being this murky diagnostic landscape,
but it turns out the x-ray isn't broken at all.
We've just been holding it up to the wrong patient.
So here's where this lands.
The idea that EVs are going to break the grid simply doesn't hold up.
Not in the data, not in long-term planning, and not in the real-world behavior of how
people actually charge their vehicles across Canada.
But the surge in energy demand from AI data centers, that's real.
And it's happening fast.
And maybe more importantly, it's happening quietly, without the same level of scrutiny
or public debate that electric vehicles have been getting.
Because the truth is, this was never just an energy story.
It's a priority story.
And once you see that, the entire narrative makes a lot more sense.
If you want more deep dives like this, cutting through noise, following the data, and asking
the uncomfortable questions, subscribe to The Sanity Project.
Because clarity is getting harder to find.
And that's exactly why we're here.
If you want more facts and less fear, hit subscribe.
Check out the next breakdown wherever you're listening or watching.
Stay sane, Canada.
Thanks & End Credits
Thanks for watching.
Check out the next breakdown.
Stay sane, Canada.
Amazon Music
Apple Podcasts
Podcast Addict
Spotify
YouTube