Most people really think about the power
grid only when the lights go out, but
the bigger warning signs right now may
come before the blackout in the rising
demand, the strained infrastructure, the
utility bills, and the summer peak loads
that are already forcing the country to
ask whether we can build power fast
enough to keep up. For nearly two
decades, electricity demand in this
country was relatively flat. Utilities,
they could plan around slow growth,
gradual upgrades, efficiency gains, and
predictable demand. And then the
forecast changed. Artificial
intelligence, data centers, electric
vehicles, manufacturing, reshoring, heat
waves, electrified homes, and modern
life have all started pulling more
aggressively on the same system at the
same time. The lights, they still may be
on in your house today, but the system
behind that switch is being asked to do
more than it was built to handle
comfortably. AI is not going to cause
every outage, and data centers are not
going to explain every increase in your
power bill. And this video is not just
about AI. There's more to this problem
than that, as I'll explain. AI is just
the clearest example right now of a much
larger shift where America is adding
more electric demand faster than the old
planning process was designed to handle.
Data centers, electric vehicles,
manufacturing, electrified homes, heat
waves, and modern life, they're all
pulling on the same system at the same
time. A simple way to think about it is
to think of an older house. Many old
houses used tape and tar to insulate
wiring. The wiring may have worked fine
for lamps, a refrigerator, and a few
basic appliances, but once you started
adding space heaters, window AC units,
freezers, power tools, and EV chargers
without upgrading the panel or wiring,
the margin disappears. And the house may
still work most days, but one hot
afternoon, one overloaded circuit, or
one more major appliance can suddenly
trip the breaker, creating a short, or
even start a fire. The electrical grid
in America is at a critical juncture
right now. And we're not talking about a
small bump in the road. US electricity
use has already been pushing into record
territory. EIA projections show total
consumption continuing to rise as homes,
businesses, data centers, factories,
vehicles, heating, cooling, and
industrial users all pull on the same
system. And for a long time, efficiency
has helped hide the problem. A newer
refrigerator used less power than the
older one. LED lights replaced
incandescent bulbs. Better building
codes helped in some places. But
efficiency can only offset so much when
the economy starts adding enormous new
loads. And we're not just talking about
more people flipping more switches. It's
an entire categories of demand that
either barely existed a generation ago,
or were nowhere near this scale. AI
training clusters, cloud computing,
battery plants, semiconductor
facilities, electric vehicle charging,
electrified heating, more cooling demand
and hotter summers, and more backup
systems, they all make sense when you
look at them one at a time. But the grid
does not feel them one at a time. It
feels a total load. For years, the power
grid had a strange advantage because
demand was not growing the way many
people expected. Homes, they became more
efficient. Light bulbs used less power.
Appliances improved. And some industrial
demand shifted. And we've covered this
in previous videos on this channel.
Utilities have had some serious
maintenance problems, aging equipment
and regional vulnerabilities. But at
least the overall demand curve was not
exploding upward every year. And those
assumptions, they really shaped decades
of planning. Power plants, they take
years to permit and build. Transmission
lines can take even longer because they
cross counties and states, private and
public land, and they require
environmental reviews and face political
opposition. Large transformers are not
the kind of thing that you can pull off
a shelf when one fails. Some of these
parts are custom built, expensive,
heavy, and slow to replace. And the grid
is not like an app that you can update
overnight. It is physical infrastructure
stretched across the country, built in
layers over generations. And electricity
has to be generated, moved over
transmission lines, stepped down through
substations, distributed through local
networks, and delivered to your house in
real time. And the grid, it has to keep
supply and demand balanced every second.
And that balance becomes harder when
demand shows up in places where the grid
was not built to handle it. Or when
generation is available on paper, but it
can't be moved easily to where it is
needed. And one failure can force power
onto other lines, and if those lines are
already stressed, the problem can
spread. And when demand was relatively
flat, utilities, they had more room to
patch, delay, plan, debate, and spread
out costs. But when demand starts rising
again, old assumptions break down. And
the question is no longer whether
America can eventually build more power,
the question is whether generation,
transmission, transformers, permitting,
labor, fuel, supply, and local
communities can move fast enough to keep
up with the demand being added right
now. And that's why today's sponsor, the
Anker Solix F3800 Plus, is a part of my
setup. One thing I've learned while
building out my property is that when
the power goes out, it's not just the
lights you have to think about. For me,
it's water. My well runs on a 240-V
pump, so reliable backup power is
important. That's one reason I've been
using the Anker Solix F3800 portable
power station. What I like about it is
that it supports both 120-V and 240-V
output, and delivers up to 6,000 W of
power from a single unit. Through the
home power panel, two units can be
combined for up to 12,000 W of output,
allowing multiple larger loads and
appliances to run at the same time.
Another standout feature is the
expandable design. Each F3800 starts at
3.84 kWh of capacity, and can be
expanded up to 26.9 kWh with additional
batteries. Pair two units together, and
that grows to as much as 53.8 kWh of
total capacity. I've also connected
solar panels to help keep it charged and
extend run time during longer outages.
Right now is actually a good time to
take a look because Anker Solix is
running its Prime Day promotion with
some of the lowest prices of the years
across its lineup. They're also
including free gifts with qualifying
purchases and a 30-day price guarantee.
If you'd like to learn more, check out
the link in the description section
below. The grid was built for one kind
of America, and now it is being asked to
power another. The numbers make that
hard to dismiss. Lawrence Berkeley
National Laboratory found that data
centers use about 4.4% of all
electricity in the United States in
2023, and by 2028, that number could
rise to somewhere between 6.7 and 12%.
The International Energy Agency projects
that global data center electricity
demand could more than double by 2030
with AI-optimized data centers driving
much of that growth. Putting that in
plain English, data centers are moving
from something most people never thought
about to something that could soon use a
share of US electricity that looks more
like a major sector of the economy. And
that's not a few server rooms tucked
away in office spaces. That is a new
layer of national demand being added on
top of homes, hospitals, factories,
schools, grocery stores, water systems,
and everything else already depending on
the grid. And this is the part that
should make people stop, pay attention,
and even sweat a little. For years, the
grid had some breathing room because
demand was mostly flat. And now some
forecasts are bending upward at the same
time utilities are still trying to
replace aging equipment, connect new
projects, build transmission, find
transformers, and prepare for hotter
summers and sharper winter peaks. The
demand curve, it is no longer drifting
upward slowly. In some regions, it is
starting to turn sharply. And that is
why this does not feel like normal
growth. Normal growth gives utilities
time to plan. And this kind of growth
forces utilities to react. And when the
system shifts from planning to reacting,
the cost lays in reliability risk and
will stay inside utility boardrooms.
Eventually, they move outward in the
communities, construction fights,
connection delays, and monthly bills.
The biggest new load in the story is
artificial intelligence and the data
centers behind it. A data center is not
just a building full of computers. At
the scale we are now talking about, some
facilities behave more like major
industrial power users. And Department
of Energy advisors have discussed
hyperscale AI facilities requiring 300
to 1,000 MW or more, which means these
are not small office buildings with
servers in the back. They are power
plant scale demand. One large AI campus
can require as much electricity as a
midsize city. And that is a mental shift
that people need to make. We're not
talking about a few office buildings
plugging into the grid. We're talking
about concentrated, round-the-clock
demand that can arrive in one region
faster than the surrounding power
infrastructure can be expanded. We're
talking about whole new cities of demand
popping up within cities that already
have demand issues. Cities of demand on
top of a city's demand. That scale
matters because a data center does not
behave like a neighborhood. It needs
reliable electricity, cooling
redundancy, and fast, around-the-clock
connectivity. And at the same time, the
business behind them is moving much
faster than the old grid planning
process was built to handle. A company
can buy land, raise capital, build
server halls, and push for connection
quickly, but the transmission lines,
substations, transformers, generation
capacity, and local approvals needed to
serve that load often move on a much
slower timeline. The pressure starts to
show up in the gap between digital speed
and physical infrastructure. AI here is
a headline, but it's not the only source
of demand. Manufacturing is part of this
because the United States is trying to
bring back more industrial capacity here
at home, especially in semiconductors,
batteries, defense production, and
advanced manufacturing. And those are
power intensive industries that really
require reliable electricity on a large
scale. Electrification is another part
of this equation. Electric vehicles,
they do not break the grid on their own,
but millions of vehicles charging over
time can become a real strain. Heat
pumps, induction cooking, electric water
heaters, and then electrified industrial
processes all change the demand profile,
too. Some of these technologies can be
managed intelligently, and some can help
balance load if planned correctly, but
they still require a grid that is ready
for them. Weather is a third-party this
equation. Recent heat events are not
just hot afternoons. In many places,
they are long stretches of high
temperatures that settle over region and
keep air conditioners running for days.
That turns cooling from a comfort issue
into a life safety issue. Sudden cold
snaps can create the same kind of stress
in the other direction, especially when
heating systems, fuel supply, and
electric demand all surge at once. And
when more of daily life depends on
electricity and the grid margin is
reduced by electrification and AI, every
major weather extreme becomes a critical
power then. And this demand is not
likely to fade quickly because it is
tied to a global competition. And here's
the thing. Artificial intelligence is
not just a side project anymore. It's
being treated as a strategic industry by
companies and countries. Closer to a new
industrial arms race rather than a
passing tech trend. And that difference,
it matters. If this were just another
speculative tech bubble, you could argue
that the demand might cool off once
investors lose interest or companies
realize the math doesn't work anymore.
But, that's not how the major players
are treating it. The United States sees
AI as an economic, military, and
technological priority. China sees it
the same way. And the largest technology
companies see it as the next platform
shift. And they don't want to be the
company that falls behind. The pressure
to build is coming from a lot of
different directions at once. It is
coming from Wall Street, Silicon Valley,
Washington, Beijing, defense
contractors, cloud companies, chip
makers, and every major business that
thinks AI may determine who wins the
next decade. Even if some individual
projects slow down, even if some
companies overbuild, and if parts of the
AI market correct, the broader race for
computing power is unlikely to simply
disappear. And that is also very
different from the crypto mining surge
we saw a few years ago. Crypto mining
can move quickly, chase cheap power, and
shut down when prices change. But AI is
becoming more deeply embedded into the
cloud computing, search, logistics,
finance, military planning, software
development, medicine, manufacturing,
and the basic tools businesses use every
day. Once companies build their
operations around that kind of
computing, the demand becomes harder to
unwind. Seen that way, the electricity
demand starts to make more sense. These
companies are building at the scale
because computing power has become
strategic power. AI may look like
software from the outside, but beneath
the surface is a physical infrastructure
race. It takes chips, servers, cooling
systems, land, water, substations,
transmission access, and enormous
amounts of electricity. The larger this
race becomes, the more the digital
economy starts competing with
households, factories, hospitals, and
local communities for the same
underlying power system. For years, many
Americans really thought of the internet
as weightless. Files were in the cloud.
Apps lived on phones. AI felt like a
software. But the cloud, it's not a
cloud. It is buildings, land, water,
transformers, substations, backup
generators, power purchase agreements,
transmission lines, workers, steel,
concrete, and electricity. The digital
world has a physical footprint, and now
that footprint is getting a lot larger.
A company announces a new AI campus, and
that campus needs a huge amount of
electricity. The utility has to figure
out how to serve it, which may require
new generation, new substations, new
transmission, or new capacity purchases.
Those projects require money, labor,
equipment, permits, and time. And if the
system can't keep up cleanly, the
pressure shows up through delays, price
increases, local conflict, or
reliability concerns. We're already
seeing the signs as large power users
wait longer to connect, utilities
rewrite their demand forecast,
communities fight over new data centers,
and grid operators warn that the power
demand they expect years from now is
starting to show up much sooner.
Building the grid is slower than
building demand. A data center can be
planned around a business cycle, but a
grid infrastructure is planned around
engineering, regulation, land rights,
supply chains, and public approval. A
new gas plant can take years. A nuclear
restart or life extension is complex. A
new transmission corridor can get tied
up for a decade. Large transformers,
they have long lead times. Skilled
electric labor is not infinite. And this
conflict is playing out in communities
across America, and maybe at this very
moment in yours as well. Right now,
local to me in Temecula, California,
SDG&E is proposing 140 mi of 500 kV high
voltage transmission towers that would
run directly through the city. And even
though this town, Temecula, doesn't get
power from this line, and the city has
officially opposed it. And your
community may be engaged in a similar
power conflict. And if you are, I'd love
to hear about it in the comments below
this video. Let us know. Even when most
people agree more power is needed, the
process is slow, and the agreement
usually breaks down over where it goes,
who pays, and who has to live next to
it. Communities they push back against
new transmission lines. They push back
against power plants. They push back
against data centers because of noise,
water use, land use, local tax deals,
and the feeling that residents carry the
burden while distant companies get the
benefit. And that local resistance is
not always irrational. People are
looking at their own roads, their water,
their own property values, their own
bills, and beginning to ask why their
community should absorb the strain of
someone else's growth and profits.
National ambition is running into local
reality. Everyone wants electricity to
be reliable. Many people want AI
progress. Many people want cleaner
energy. A lot of people want
manufacturing brought back to the United
States.
People want electric vehicles, heat
pumps, battery plants, chip factories,
and modern infrastructure. But, all of
those goals, they require power. Not
symbolic power or political talking
points, but actual electricity delivered
to the right place at the right time
through physical infrastructure that
already has stress points.
And the difficulty in this is a
collision of all these forces. There is
not one villain, one policy, or just one
company behind it all. The system is
being asked to do more at the same time
that replacing and expanding it has
become expensive, slow, and politically
difficult. Weather is where all of this
pressure gets tested in real time. And
the grid may look stable on an average
day, but average days are not usually
what expose the weakness. The stress
shows up when heat settles over a region
for several days and millions of air
conditioners run almost non-stop. Or
when a sudden cold snap forces homes,
businesses, and utilities to pull hard
at the same time. Add a storm, wildfire,
flood, high winds, a planned outage, or
a fuel constraint into the same window,
and guess what? The grid has less room
to absorb the surprise. The danger is
not that one hot day or extreme weather
event automatically takes the country
down, although we've seen that kind of
thing in regions of the country. The
concern is that a system with less
cushion has fewer ways to recover when
several things go wrong at once.
Northern Virginia is one of the clearest
examples of how this becomes more than a
technology story. It became the center
of the data center world because of
fiber connections, land availability,
tax policy, proximity to major
customers, and an existing technology
corridor. But, concentration creates
strain. If a region becomes the place
where the digital economy wants to plug
in, the local grid has to absorb that
demand. So, eventually those costs leave
the planning documents and move into
utility filings, capacity markets, local
politics, and the question every
household cares about, which is who
pays. In Virginia, state-level analysis
has warned that ordinary residential
customers could eventually see higher
monthly bills if the cost of new grid
infrastructure is spread across everyone
instead of being isolated to the largest
new users. In the PJM grid region, which
serves tens of millions of people across
the Mid-Atlantic and Midwest, the market
monitor has warned that data center
demand is already contributing to sharp
increases in capacity prices. Wholesale
market costs are not the same thing as
your home bill, but they show how grid
strain can move through the system. Now,
demand requires more generation, more
transmission, and more backup capacity.
And those costs, they show up in utility
plans, capacity markets, future rate
increases, and long-term infrastructure
charges. And over time, they can work
their way into the retail bill that
lands, guess where? At your house. The
average American may never step inside
an AI data center, but they may still
help pay to power one. Most people hear
AI data center and they think it has
nothing to do with them, but when a new
data center requires substations,
transmission lines, backup capacity,
generation, or other grid upgrades,
those costs, they don't disappear. The
corporation may pay for part of it, but
it often does not pay for everything
needed to strengthen the surrounding
grid. So, in plain English, households
can end up helping fund the upgrades
needed to serve massive new users, even
if those households never even ask for
the project and only notice it when the
monthly bill gets harder to absorb. Seen
from the household level, the issue is
not only whether the lights go out, it
is price, priority, and margin. Someone
has to decide who gets connected first,
who pays for upgrades, which communities
carry the water, land, and power burden,
which households see higher bills, which
regions have enough reserve capacity
during heat waves, and which utilities
are already falling behind. Households
can feel that in very practical ways. A
few hours without power can be
inconvenient, but a day without power
can ruin refrigerate food, shut down
work, knock out the internet, and make a
house dangerous in extreme heat or cold.
Several days without power can affect
water, medical devices, communications,
transportation, fuel, and local stores.
Rising electricity costs create a much
different kind of strain because they
hit every month. They affect
fixed-income households, families
already stretched by food, insurance,
rent, and debt, and small businesses
that need refrigeration, lighting
equipment, or climate control. A grid
can remain mostly functional and still
become harder for ordinary households to
live under. Grid operators live in the
world of margins and they think about
reserves, peak demand, available
generation, transmission constraints,
and what happens if a large generator
trips offline at the wrong moment. Most
households do not think that way because
they don't have to. They only know
whether the power works. They know the
grid has a problem when the lights go
out or when they flip a switch and
nothing happens. So, when demand grows
faster than expected, reserve margins,
they become important. A region can look
fine most days and still be vulnerable
during a heat wave. It can have enough
annual energy and still lack enough
deliverable power at a specific hour in
a specific place. And it can have
generation on paper that does not help
if transmission is constrained. And that
is the part that I think a lot of people
are missing here. The issue is not just
whether America can produce more
electricity in a general sense. The
issue is whether the right electricity
can reach the right place at the exact
moment that most people need it. The
answer is not as simple as saying
America needs more power plants. America
needs the right power in the right place
with the right transmission at the right
time and enough backup and flexibility
to handle peak stress.
That problem is much harder than a
slogan because every part of the
solution takes time. New generation has
to be financed, permitted, built,
connected, and fueled. Transmission
lines have to cross rural communities,
and transformer substations and
switchgear may have to be manufactured
and installed. Regulators have to
approve the cost and someone eventually
has to pay for them. So, what happens
next probably will not be one dramatic
national blackout that reveals a whole
problem at once.
It will likely show up unevenly with
faster progress in some regions and
slower progress in others. Some
utilities are going to buy themselves
more room by managing demand, building
capacity, and upgrading infrastructure
early. Others are going to still
struggle because demand arrives faster
than the planning process or equipment
or their local politics can absorb. Data
center states may attract investment and
jobs, then spend years arguing over who
has to pay for the power lines,
substations, backup capacity, and water
demand needed to support them. And
that's why the next few years are going
to feel confusing. You may be already
seeing headlines about new power plants,
nuclear restarts, large battery
projects, transmission corridors, smart
grid upgrades, and massive investment in
energy infrastructure. At the same time,
you're going to also see higher bills,
project delays, local opposition,
capacity warnings, and more concern
about peak demand. Now, both can be true
at once. The country can be building
more power and still feel strain while
trying to catch up. For households, the
outcome may look less like one national
crisis and more like a steady loss of
margin. It may show up through higher
utility bills, more pressure during peak
demand, more fights over local
infrastructure, and the outages that
still come from storms, wildfires, heat
waves, equipment failures, or human
error. But, it's going to land harder
because the system underneath has less
room to absorb the shock. The common
thread through all of this is margin. If
demand keeps rising while infrastructure
lags, that margin is going to get
thinner. Disruption begins to matter
more because the system has less room to
absorb them. None of this is a reason to
panic, but it is a reason to stop
treating electricity as a guaranteed
background service that's always going
to be cheap, stable, and available
exactly when you need it. For your
household, the first step is not buying
the biggest generator that you can find.
The first step is deciding what level of
disruption you want to be able to
ignore. A 6-hour outage, an overnight
outage, a 2-day outage, a recurring
outage, they're all different problems
and they require different levels of
preparation. And that is where a lot of
people get stuck. They start with
products and immediately fall into the
rabbit hole of solar generators, gas
generators, whole home batteries,
rooftop solar, transfer switches, and
large battery systems. Before long,
they're buried in specs, wattage charts,
runtime estimates, reviews, opinions,
and endless debates over which setup is
best. After a while, all those options
can turn into paralysis. Look, don't get
me wrong. Those details they do matter,
but they come after the most important
decision, which is what your household
actually needs to keep working when the
power's out. A practical power plan does
not require you to disconnect from the
grid and it does not require you to jump
straight into a full energy
independence. Most households can still
reduce their exposure one layer at a
time by identifying critical loads,
protecting refrigeration, keeping
communications running, planning for
medical needs, and creating at least one
way to function when the grid is under
strain. The country may eventually build
enough generation, transmission,
storage, and backup capacity to handle
this new era of demand. AI
infrastructure may also become more
efficient and I hope both happen, but
they're going to take time, money, and
political will and your household still
has to live through this transition.
I'm probably like you in that I don't
want to hang the stability of my home on
hope or leave it entirely in the hands
of tech companies, utilities, and
politicians. In part two of this video
series, we're going to walk you through
the levels of household power resilience
and how to think through them without
getting overwhelmed or wasting money on
the wrong setup. I'm going to link to
that video below when it's available,
also along with the city prepping brief
and any backup power guides that we
built for you because this channel is
not just about identifying problems,
it's about finding practical solutions
where they matter most at home. As
always, stay safe out there.