You have an incredible amount of power.
Why [music] should we trust you?
>> Um, you shouldn't trust me.
>> I don't think most people understand
just how quickly
>> [music]
>> machine intelligence is advancing.
>> If this technology goes wrong, it can go
quite wrong. I think AI will probably,
like most likely, sort of lead to the
end of the world, but in the meantime,
uh, there will be great companies.
>> AI could change almost everything, and
that's exactly what makes figuring out
how to prepare for it so frustrating.
When people hear artificial
intelligence, they may be thinking about
completely different problems. Jobs are
probably the most obvious fear. And then
there are deep fakes and voice cloning,
scams that are getting harder to
recognize, and cyber attacks that can
move faster and more effective. Beyond
that, are concerns about banking,
communications, critical infrastructure,
and eventually much more autonomous
systems. And of course, there's a
version that we've all seen in the
movies where the machines eventually
decide that we are the problem. And that
is the huge range of possibilities to
put under one label, which is part of
why AI still feels so abstract. But the
technology is moving quickly. Its
capabilities they keep changing, and I
don't think anybody can confidently map
out all of the downstream consequences
5, 10, or 20 years from now. There are
laws, security standards, regulators,
and companies that are trying to build
safeguards around it, but none of that
can protect every part of your life that
AI may affect. So rather than waiting
for the picture to become perfectly
clear, I think it makes sense to start
building some resilience now. Before we
get into the practical steps though, I
do want to give you a simple way to
think about the problem. Otherwise, this
really turns into another checklist
where I tell you what to do without
really explaining what you're preparing
for. Now, the easiest way I found is to
organize it by time. Some effects are
already here. Others are clearly
accelerating, and then there's a third
group where the consequences could be
much larger, but they're still genuinely
uncertain. So let's jump in. If you're
new to this channel, my name is Chris
and on this channel we discuss emergency
preparedness, aka prepping. So, let's
start with this explanation with what is
already happening. AI is being used in
scams, impersonation, phishing, and
other cybercrime. Fake voices, images,
and videos are already getting more
convincing while workplaces are already
changing how certain tasks are done.
Now, the next horizon is what seems to
be accelerating. Businesses, they're
adopting these tools really quickly and
the systems themselves are getting
better at handling longer, more
complicated task. At the same time,
synthetic content keeps getting cheaper
to produce and AI is being built into
more of the digital systems that we just
interact with every day. And then,
there's the uncertain horizon. We don't
know how many jobs ultimately disappear
or how disruptive the economic
transition becomes. And the same
uncertainty applies to more autonomous
systems and the possibility they create
serious problems for infrastructure or
other services that we depend on.
Now, those questions they matter, but
pretending that we know the answers
would really not help you prepare. What
does help is narrowing the problem.
Instead of just treating AI as one
enormous threat, look at the parts of
life it can make less reliable. Your
income, your identity, the information
that you trust, your money and online
accounts, and the digital systems your
household depends on.
Once you do that, this starts to look a
lot like a normal preparedness problem.
And before we jump into the first one,
if you want something practical you can
use to start protecting your household,
we have updated our free cyber attack
guide. It now includes a new section on
AI-enabled scams and digital threats
along with steps to protect your
accounts and to prepare for disruptions.
I'll put a free download link below.
Number one, your job and your income.
So, let's start with probably the
biggest concern that most people have.
Is AI going to take my job?
Nobody knows exactly how large the
disruption is going to become, but the
concern is no longer theoretical. A Pew
Research Center survey conducted in June
found that 71% of US adults think AI
will lead to fewer jobs in the United
States over the next 20 years. Now, that
number is not saying 71% of Americans
are going to lose their jobs. It just
tells us how widespread the expectation
of disruption has become. The labor
research, it gives a much more
complicated picture. The International
Labor Organization estimates that
roughly one in four workers worldwide is
in an occupation with at least some
exposure to generative AI. But, exposure
is not the same thing as replacement.
And for most of those occupations, the
more likely outcome is that the work's
going to change. Some tasks they'll
become automated, uh certain roles may
need fewer people, and the value of
particular skills shifts. Now, how
quickly that happens is going to vary a
lot from one industry to another. But,
for me that changes the preparedness
question. I would spend less time trying
to predict whether an entire profession
disappears and spend more time looking
at how AI is entering your field right
now. You want to pay attention to what
companies are using it for, which parts
of your job are getting easier to
automate, and where human experience,
judgment, accountability, uh
relationships, physical skill, or
specialized knowledge, they still
matter. What that does is it gives you a
much better picture of where you may
need to adapt. Ignoring this, uh it's
probably a mistake. Now, you don't have
to love AI, and you should not hand over
your judgment over to it, but if those
tools are moving into your industry,
understanding them, it gives you
options. So, learn the ones that are
relevant to your work and figure out
where they can make you more capable.
And at the same time, keep developing
the skills that still matter when the
tool is wrong, unavailable, or just
simply not enough. And then, there's a
household side of the problem, financial
runway. If your industry changed faster
than expected, how long could your
household absorb the disruption? And
that's why emergency savings still
matter here, just as they do during a
recession or a layoff. And if most of
your income depends on one narrow skill
set or a single source of income,
developing another useful skill or
possible income stream, it gives you a
lot more room to adjust instead of
having to react under pressure. And you
don't need a perfect forecast of the job
market. The goal is simply to have more
options if your part of it changes
faster than you expected. And that idea
is going to repeat throughout this
video. Preparedness, it's less about
predicting the exact future and more
about making sure that one change does
not take away every path forward.
Two, your identity. Who can you actually
trust?
Now, let's move from a risk that is
difficult to forecast to one that's
already showing up in the real world.
AI-assisted fraud and impersonation. The
FBI's 2025 Internet Crime Report logged
more than 22,000 complaints involving
AI-related information, representing
more than 893 million in adjusted
losses. Now, that doesn't mean that
every dollar came from a cloned voice or
a fake video. AI is used across
different kinds of fraud, but it does
point to a basic household problem that
we need to get used to. Seeing or
hearing someone is becoming weaker proof
that it is really them. We are used to
treating familiarity as authentication.
If your caller sounds like your son, you
assume it's your son. If a video appears
to show someone you know, your first
instinct is to trust what you're seeing.
And AI weakens that shortcut. The FBI
has already warned about campaigns using
AI-generated voice messages to
impersonate people. And as the
technology improves, trying to teach
everyone in your family to spot every
fake, I think that's really a losing
strategy. But a better approach is to
agree on a verification process before
anyone needs it.
Uh family word or a question, it can be
part of that. But the real protection is
a habit behind it. Unexpected requests
involving money, passwords, account
information, or an emergency, they don't
get acted on until someone verifies them
through the channel the family or a
trust.
Here's a scenario. Imagine getting a
call from your daughter. She's crying.
She says she's been in an accident and
she needs money immediately. Now, the
voice may sound completely convincing,
and instead of trying to decide whether
the audio is fake while you're scared or
under pressure, end the call and contact
her through the number already stored in
your phone.
If she doesn't answer, reach someone who
should know where she is. But the point
here is to get outside of the
communication channel the scammer
controls. And that same habit works
beyond family scams. A strange request
from your boss, it deserves a second
channel of verification. And a bank or a
government message should send you to a
number, an app, or website you already
know is legitimate. The more urgency
someone tries to create, the more reason
you have to slow down the process. And
talk through that rule with children and
older relatives ahead of time because
criminals, they're still exploiting the
same old human vulnerability.
Fear makes people act quickly. AI just
makes a story easier to believe. Now,
once identity becomes harder to verify,
the next problem follows naturally. What
happens when the information itself
looks convincing?
Three, information. What can you
actually trust?
A convincing image, an audio clip, a
screenshot, or a video is no longer
enough by itself to prove that something
happened. That matters all the time, but
I think it becomes especially important
during emergencies when people are
moving quickly and bad information can
spread before anyone has time to check
it. Trying to turn yourself into a
deepfake expert is probably not the
answer. These systems, they are
improving, and even detection tools can
struggle outside controlled testing. A
better habit is to shift the question
from does this look real to where did
this come from, and can I verify it
somewhere I already trust? Now, during
an emergency, that might mean checking
an evacuation order with the agency that
issued or going directly to the utility
if there's a warning about your water or
power. A scary bank message should
really push you toward the bank's own
app or website, not the link in the
message. And when a dramatic video
starts racing around social media,
especially during a disaster, sometimes
the smartest move is just simply to wait
long enough for credible sources to
confirm what actually happened. This is
also where old school preparedness is
earns its place. You want to keep
important phone numbers somewhere other
than your phone, know which local
agencies and utilities you would turn to
during an emergency, and have another
way to receive information if your
normal apps or internet connection are
unavailable. Now, the technology, it may
be changing, but the principle is
familiar. When the information matters,
verify the source before you act.
Four, your money, accounts, and
recovery.
Now, your digital accounts are another
place where a small failure can spread
much farther than you expect. AI can
make fishing, social engineering, and
other cyber attacks faster and more
convincing, but the household
vulnerability often has less to do with
AI itself and more to do with how much
of our lives really now funnel through a
few access points. Think about your main
email account and your phone. One may be
the recovery path for your bank, your
insurance, your cloud storage,
utilities, and dozens of other services,
while the other receives authentication
codes for many of the same accounts. If
both become unavailable tonight, the
real problem would not be losing two
devices or logins. It would really be
about discovering how many other doors
they unlock. And that's where I would
focus the security work. Unique
passwords or passkeys, they make it
harder for one compromised account to
lead into another. And multi-factor
authentication adds another barrier.
Now, on the accounts that matter most,
fishing-resistant options, they're
really worth using when they are
available. Recovery settings, they
deserve the same attention. So, make
sure the phone numbers and email
addresses are current and keep recovery
codes somewhere secure that does not
depend on the same phone number or cloud
account that you may be trying to
recover. Now, there's also a simple
continuity question that most families
really never think about until something
goes wrong. How much does one person
know that nobody else does?
Another trusted adult may not need a
folder full of every password, but they
should be able to figure out how
essential bills get paid, where the
insurance information lives, how to find
critical records, and who to contact if
the person who normally handles all of
that is unavailable.
Payments also need some redundancy as
well. A second card or an account along
with a reasonable amount of emergency
cash really gives you another path when
the normal one fails. And I'm not
predicting an AI-driven banking collapse
here. Ordinary outages, locked accounts,
lost phones, or payment problems, they
are enough to make the backup useful,
especially when you're trying to buy
food, fuel, or medicine.
Five, audit your digital dependence.
Now, one of the most useful exercises
you can do after watching this video
really doesn't have anything to do with
predicting what AI becomes. I would
encourage you to walk through your home
and notice how many important functions
depend on an app, a cloud account, an
internet connection, or someone else's
server working the way that it's
supposed to. For one household, that
might be cameras and smart locks.
Somewhere else, it could be solar or
battery equipment, maybe an EV charger,
medical information, cloud documents, or
equipment that's tied to a property
system. I just read a report about
Samsung's AI refrigerators that stopped
working after a bad software update. And
that was not an AI failure, but it is a
good example of the larger issue. We are
steadily building more digital
dependencies into our normal life, and
AI is becoming another layer inside that
environment. So, pick the systems that
would actually matter if they stopped
working and learn what the fallback
looks like. Some are going to keep
working locally, others may have a
manual override or instructions that you
should have offline. And that same
exercise applies to information and
communication. Know what you can still
access without a cloud login and what
your household would do if the internet
or payment network was temporarily
unavailable. You're looking for the
places where one digital failure creates
a much larger practical problem. And
there's no reason to rip smart devices
out of your house or return to doing
everything on paper. I mean, I use
plenty of technology myself, but the
goal is simply to notice the single
points of failure while you still have
time to build around them. If a system
matters, know your alternate way to
operate it, access information, or get
through a temporary outage.
Six, learn to use AI without outsourcing
your judgment.
Preparedness it also means learning how
to use a technology that's changing the
environment around you.
I use AI and I think refusing to learn
it is going to create problems for a lot
of people. It can help with research,
organization, troubleshooting, learning
a new subject, comparing options, and
getting certain kinds of work done
faster. If these tools are moving into
your profession or industry,
understanding what they can and cannot
do is a lot more useful than trying to
pretend that they're not there. Where I
would draw the line is judgment. A
confident answer can still be wrong and
there are different AI services that
handle data differently. And before
sharing sensitive financial, medical,
legal, business, or personal
information, I would encourage you to
understand how that particular service
uses and stores what you give it. When
the consequences are serious, whether
that involves health, money, electrical
work, machinery, water, or a legal
decision, I would encourage you to use
AI as a starting point and then verify
the parts that matter with an
authoritative source, a manual, or a
qualified professional. And the goal is
to let the tool make you more capable
without becoming helpless when it's
wrong, unavailable, or simply not the
right tool for the decision.
I still can't tell you exactly where all
of this is going and I don't think
anyone really can. But we don't need to
know exactly where it's going to prepare
for it now. What really matters is
noticing where AI could make something
that your household depends on less
reliable, whether that is your income,
your identity, the information you
trust, your money, or the digital
systems that are around you. And then
asking whether you have another path if
that happens. Again, we provide you with
a free download guide that I'm going to
recommend that you check out in the
description and pinned comment below.
You also do not need to turn this into a
weekend project where you try to fix
everything at once. Maybe the most
obvious weakness in your house is that
nobody has a verification plan for an
emergency call. For someone else it may
be account recovery, important records
that exist only in the cloud, or having
no backup way to pay. Your job may be
the bigger concern, in which case
spending some time understanding how AI
is already being used in your field
could be the first most useful step for
you. The point is to choose the
vulnerability that really matters most
in your household and work on actually
improving it.
I'd like to hear where you land on this.
When you think about AI in your
household, what concerns you most right
now?
It could be your job and income, scams
and identity theft, misinformation,
cyber attacks and banking, or our
growing dependence on digital systems.
Maybe it's your job and income, scams
and identity theft, misinformation,
cyber attacks and banking, or our
growing dependence on digital systems.
Of all those things I just listed, I
would say they're all concerns for me.
So, let me know what I missed in the
comments below. And also be sure to
check out a couple of videos I'm going
to post here on the side of the screen
and in the description and comment
section below. The first video is what I
learned about AI is changing my plans,
and the second video is 18 survival
items to get before 2027. They're going
to help you understand where things are
headed with AI, and they're going to
give you practical steps that you can
take to prepare. As always, stay safe
out there.