All language subtitles for [English (auto-generated)] They’re Watching You Through Wi-Fi… And You Have No Idea [DownSub.com]

Almost everybody in the industrialized

world now has Wi-Fi as part of their

home, their office, what have you. The

combination of radio technology and

cyber security and AI all come together

to create something that we never

thought we would be able to do. This is

the same thing, sending out the Wi-Fi

signal and then getting the signal and

having enough intelligence built into

your system to be able to determine what

is going on in that room.

Hey everyone, Yaniv Hoffman back with

another video and back with the one and

only Master Aer Occupy the web. Welcome.

>> Thanks Yaniv. It's good to be back.

Thank you for the honor of hosting me

again. I really appreciate that.

>> Yes. Yes. Already got complaints and

questions. Where is OTW? So happy you

are with me. And today we have a a cool

I believe a very cool subject. And what

if I told you that your Wi-Fi can see

you even through walls. So a new

technique lets attackers track your

every move inside your home without ever

stepping inside. And to break it down,

I'm turning it to OTW, who is going to

show you exactly how this works. Well, I

don't know if I'm going to show them

exactly, but I I will describe

exactly how it works, right? So, there's

almost everybody in the industrialized

world now has Wi-Fi as part of their

home, their office, what have you. And

we've just kind of taken that for

granted. We've now entered the radio

signal signals intelligence era. We are

entering the signals intelligence era of

cyber security. And this new technology

is an indication. It's a it's a

manifestation.

of the combination of radio

technology

and cyber security and AI all come

together to create something that we

never thought we would be able to do.

And what it is is it's the ability to

send a Wi-Fi signal. Now Wi-Fi is simply

a radio signal. That's all it is. You

know, like so many things around us,

whether it be your cell phone or your

microwave oven or even the light that's

around you. These are all

electromagnetic magnetic waves. Wi-Fi is

no different. It just is a way of using

radio signals to transfer digital

information. Okay? That's all it really

is. Now we know that when there are

radio signals or for that matter sound

or light when it hits something right

the the electromagnetic wave is

scattered or refracted or absorbed.

Those are the three things that can

happen. Well, if it's absorbed, then you

know, we don't really know what happened

to it. But if it's refracted or

reflected, we can it comes back to us.

All right? It's like it's like eyesight.

You know, eyesight is simply the light

going from a it's reflecting off a

surface and coming back to your eye.

This is the same thing sending out the

Wi-Fi signal and then getting the signal

and having enough intelligence

built into your system to be able to

determine

what is going on in that room. All

right, you can by simply seeing the

refraction and the bouncing of the radio

waves. Now, it's been something that's

been around, the idea has been around

for at least 10 years, maybe longer,

right? But right now, with the advent of

AI in cyber security and the advent of

AI in what I call signals intelligence,

right,

we can now build one of these devices

that can see through walls, right, for

relatively small amount. We're not

talking about military budgets. We're

talking about a home hobbyist budget.

The the biggest expense in doing this is

you have to have video card. You have to

have a video, right? The software is to

run on CUDA. So, so here's the system,

right? The system simply is reading the

the bouncing and refracting radio waves

and then with that information is able

to determine what is okay and what is a

wall, what's a piece of furniture,

what's animate, what's not animate and

then with that information can actually

display the outlines of the body as it's

moving in real time. the this the

scientists who developed this, the

researchers who developed it tried to

pass it off as a privacy.

It's a privacy issue. It's like no, this

is an anti- privacy issue.

It's a it's a spy device, right? That's

what it is. It's not it's not a privacy

thing. It's a spy device. It's

anti-privacy. And so

the the important point here is that it

will be used by law enforcement. It will

be used by military intelligence. It

will be used by bad guys. Right? That's

the truth with no matter what technology

you're talking about. And so it's

important I think to one to understand

what what's going on here, right? What

is actually happening? what's possible,

right? And possibly how you can protect

yourself from this type of spy,

>> right?

>> And so at at Hackers Rise, we're we're

working on building one of these devices

and we will hopefully demonstrate it

soon uh and have a course on it as well.

But yeah, it's it's basically what this

is. We we need something to send and

receive radio waves. That's what we

need, right? And that's what these

lowcost SDRs are. Okay? The cheapest

ones can't send a signal, but they can

receive a signal. And so, in some cases,

if you're using the Wi-Fi that's already

in the room, then that's adequate,

right? Because the Wi-Fi is sending out

the radio signal. You're just receiving

the radio signal. But it's also possible

to send the Wi-Fi out, you know, not

using it for any purpose other than

trying to see what's inside the room to

send a powerful signal through the wall.

Okay? It strikes the objects in the room

and then reflects back to an AI that can

read that data and reconstruct figures

in real time. Right? So that's what is

necessary to do it and you need a Nvidia

card

and you need a piece of software, right?

It's called Dense Pose. I will uh I'll

put it up on the screen for you. Okay,

this is Dense Pose, right? It was a it's

it's basically a an AI system, right?

that takes signals and converts them

into

body forms. All right? And and so this

is a it's all open source. And so this

is the software end of it. The hardware

end is there's two elements. You got to

have a computer obviously. Then you have

to have an SDR, right, to be able to

send and receive the signals.

>> And then you need um this software,

right? And once you have that, then it's

possible to build a device that can see

through walls. Now, there's some

limitations of a device like this. And

one of the limitations is that Wi-Fi as

it's specified throughout the

industrialized world is limited to

between a half a watt to one watt.

Right?

And so that's why we have ranges,

limited ranges of Wi-Fi that are only

like a 100 meters, right? So Wi-Fi in

the US is limited to a half a watt by

the FCC. But that doesn't mean that

somebody couldn't build a Wii device

that even put out more power, had a

directional antenna. So, it's putting

out the Wi-Fi at a higher power in one

direction. Its ability to see it through

walls will be enhanced,

right? It can see further than the 100

meters. There's a lot of things that we

need to consider in this, right? And

that this can be used for malicious and

nefarious purposes. And it certainly,

you know, it has all of the makings of a

2025 spy device.

>> Spy device.

>> It It definitely looks like a spy device

to me.

Uh, but I should tell you that there's a

commercial company out there that makes

home security systems based upon this.

And so what they've done is that they've

built a home security system based on

this. And then they can determine if

there is a motion within the room that

shouldn't be there, right? It takes like

a fingerprint of all of the people and

devices in the room. And so if a new

person comes in, it can tell that

there's a new person in the room. If the

a dog comes into the room, it knows it's

a dog.

If a picture falls off the wall, it's

not an emergency. It it's can to be that

intelligent, right? And so this is the

first commercial application I've seen

of it is security system. So the

original researchers were saying, well,

this is going to be a privacy device.

You know, I can't buy that at all. It's

not a privacy device. It's an it's an

anti-privacy device. But there are some

commercial applications that aren't spy

devices like home security systems,

right? That'd be one application.

Another application would be using it to

monitor old people, right? So using

Wi-Fi, you could create a picture of

their face, okay? and their breathing

and then say okay this is this person's

doing okay because the breathing is you

know is deep and it's at a certain rate

per minute and if it varies from that

then we have an emergency and we need to

take care of this. So these are ways

that it can be used um in commercial

applications.

Um and of course you know we're talking

about wifi here. All right. So Wi-Fi is

that is that everywhere but it could be

used other not just Wi-Fi. This

technology is not limited by Wi-Fi.

Right. Wi-Fi is just the most convenient

and it's in every home. Well I should

put it. It's in every industrialized

nearly every industrialized um

>> hall industrialized country's homes. So

it can be used for that purpose. You

know the the attacker doesn't have to

bring in their own signal. The signal is

being generated inside your home and all

they're doing is reading the signal.

Yeah. And that gives them access to

everything that's going on inside of

that room.

>> Yeah. Now, it could be much more

nefarious and use different signals, a

higher um frequency signal. A higher

frequency is going to get you much

greater resolution. So, the the images

that you'll see that have been generated

with this technology, they're nice human

figures. You can tell that they're

human, but you can't really tell facial

features and what have you. with with

higher frequency signals, you would get

really good resolution. So that once

again, this could be used as a spy

weapon that can see through walls.

Maybe, you know, law enforcement uses it

to be able to see who's inside the house

before they break in or what. Yeah,

>> it's kind of like it's kind of like an

X-ray.

>> Okay. But it's much more it has much

more um a resolution than an X-ray does

and it's it's it's more like an MRI

might be the best way to think of it.

It's more like an MRI

of a room. They're taking an MRI of a

room and and and basically seeing how

the radio are refracted or reflect.

That's what an MRI does too, by the way.

But it's the same electromagnetic

technology

um but much more powerful, much more

sensitive. What we're talking about is

offtheshelf

parts, a regular computer, an Nvidia

graphics card, and some open source

software.

>> You have yourself you have yourself a

spy device.

>> Yeah. So, so OTW, if I'm thinking it, I

work. So, we know that um WiFi signals

don't stop at the walls. They penetrate

they reflect bounce of objects including

human. So say when a human is moves in a

room their body disrupt the Wi-Fi signal

and in a predictable way let's say and

with the right tools like you say an SDR

of course a computer Nvidia um uh card

you can capture those distortions and

reconstruct I don't know 2D 3D

representation of the 3D 3D

>> yes 3D

Yep. This this is 3D software and it

does require the Nvidia card because the

Nvidia card has built into it CUDA and

for those for let's maybe back up a

little bit. People know Nvidia, right? I

think now of the AI company, right? They

make the CPUs, the GPUs that are used

for AI, but they've been they've been

around for a long time for oh, let's see

22

years, I believe. 22 years they've been

around and during those 22 years they

basically have been making GPUs,

graphics processing units, right, that

are a little bit different than CPUs.

They're they're basically they're

simpler devices that basically do

calculations and they move polygons very

very rapidly, right? And they've made a

business out of that for years and years

and years. They've done very well for

themselves. But the reason they've done

so well for themselves is because they

developed essentially an operating

system called CUDA, right? And a lot of

the AI the AI applications run on CUDA.

This is an example here. Okay, this one

we're talking about there's another one

that dens developed on Nvidia because of

CUDA, right? And this is why this is why

Nvidia is the leader in in making I

guess they're almost a they're four

trillion dollar company. They're bigger

than a lot of countries and and the

reason is because they're make they made

these these GPUs with CUDA and this

application requires that right this is

why they're doing so well. Of course,

the higherend applications

um require entire server farms full of

GPUs

and enough power to power an entire

city. That's why we're starting to see

um power plant uh nuclear power plants

come back online in the United States,

the ones that have been retired. And

we're going to see this other places too

because one of the things that's that I

don't know that the general public is

thinking about but the AI revolution is

going to suck up a huge amount of energy

a huge amount of energy and and so

we have to see what does how impact

that's going to have on on nations and

energy security. I mean it might be that

the AI uses you know all the energy

and leaves and leaves us with none.

Microsoft a few months back bought

themselves a nuclear power plant. That's

how much energy is being used. They

bought themselves a nuclear power plant.

But I'm not one to denigrate AI. I've

been waiting for AI all my life.

Literally all my life. like when's AI

gonna come? When's AI gonna come? And

it's here. And I think that we all need

to embrace that it's here. Um, and I

think as far as cyber security jobs,

which is a question I get all the time,

the answer is learn to integrate AI into

your work. Okay? Learn to integrate AI

into what you do every day. And if you

do that, you you're not going to be out

of work. You're gonna have a job. You're

gonna have lots of jobs. You're gonna

have high paying jobs.

>> The person the person who doesn't do

that is may be unemployed, right? They

may be unemployed if they can't

integrate AI into everything that they

do. Here's just excellent. I just love

this this device, this technology,

because it it combines two of my

favorite subjects, which is radio

hacking

with AI. And so what could get better

than that, right? AI and radio hacking.

So, so let's, you know, look for this

from us. I I wrote a little article on

hackers rise a week ago or so and hard

to explain what's involved and we'll be

pursuing this in the very near future.

>> So, so OTW, I have a question from

defense point of view because now as you

said, this is a spy device, right?

Everyone can uh do it. So what are the

counter measures that people or

organization can do to block to block

such attempts is jamming is is one of

the tactics or it will not help in this

uh

>> well let's let's talk about simple the

simple measures first

you know the the the if you were trying

to have your own Wi-Fi signal not used

could turn it down you could turn down

the power on your Wi-Fi. That's not

convenient, though. Nobody likes that

because then I can't reach it in in

different parts of the office or the

home or the school or wherever you're

at. But that's an obvious possible uh

solution. Of course, you could also use

um Ethernet, just skip the Wi-Fi, right?

and and so far

this technology has only been refined.

It's not it's possible with any radio

signal, but you could also use Bluetooth

instead of Wi-Fi. I mean, Bluetooth is

quite as fast, but it'll it'll do,

right? So, like for instance, you could

the way you tether your phone by

Bluetooth, right? You can do that. You

can send those signals and data that

way.

And then you could also try to block the

signal, right? And that would be to put

some dense objects between the outer

wall and you, whatever that happened to

be. I mean, if it's a if it's a 10-in

concrete wall, there's probably not

going to be any signals emanating

through it, right? But we can't all do

those things, right? For a lot of

people, just those are not um practical

ways. I mean you could always put in you

could always you know enclose your

entire house in lead

you know that's as you know a lot of the

highly confidential facilities in the

military and is they have lead line

walls all the way around right bec

because why do they do that doesn't

allow any signals to emanate out through

that you can't get a cell phone signal

out you can't get Wi-Fi out. Once again,

you know, if you were in a in a a state

sponsored type of of circumstance,

that's what you would do. You would go

ahead and build yourself a facility

where there was very dense material with

lead being the densest, right? Is build

with leadline walls so that the signal

can't leave. But, you know, that's for

state sponsored. That's not for you and

I. you know, the best the best thing

that we might be able to do is turn our

signal down

>> or put large objects there.

>> Um, the other thing you mentioned is is

basically jamming. And jamming is a

possibility. It's not easy. It's

something that we do in our advanced SDR

class is we we work at jamming different

signals. And one of the problems with

jamming wifi and in Bluetooth is that

they use a um a spectrum skipping

technology. So they go and they skip

through various frequencies. They send a

message on one frequency. That's the

beginning of the message. And then the

second part of the message goes into a

second frequency. And then the third

part of the message goes to a third

frequency. So you have to know what the

algorithm is that's determining what

frequency the message is on. Okay? And

that's what's called spread spectrum

technology. And that is a good strategy.

But if I can figure out the algorithm,

okay, if I can figure out the algorithm

you're using, then I can break that type

of encryption. the the pro the the issue

with say Bluetooth low energy is it's

pretty easy to break it right and we can

break it Wi-Fi is a little bit more

difficult because there is an element of

randomness okay in the algorithm and so

predicting that randomness becomes very

difficult by definition predicting

randomness is difficult

you you can't predict randomness

>> so okay W you you mentioned that you are

working on such a device while you were

talking

presented here on the screen the the

article you wrote in a rise and we'll

put it now again plus we did multiple

videos on SDR eching which are now on

the screen as well but are you going to

build a course around it?

>> Yes, we're going to build a course

around this. Um, we have a Wi-Fi hacking

course coming up uh in a couple weeks

probably by the time this video is

showing and we're going to include

elements of this in that course. We

won't actually build the device, but we

will teach how it works and how to build

it. Not actually because there's too we

have too many areas to cover in the

Wi-Fi hacking course. And we're also

going to be covering the nearest

neighbor attack. Um, that's the attack

that the Russians used to attack inside

of the United States where they went and

compromised a system in Washington DC

and then use the systems Wi-Fi to attack

the neighbor, right? They attacked the

neighbor that way to get some

information uh from the US government.

So we'll be working on that as well as

you know WPA3 and WPA2

and all the different types of we attack

the old old web. So we got a lot of

material we're trying to cover in terms

of you know evil twin attacks and rogue

access attacks. So rogue APS. So we're

going to introduce this that class and

show people a bit about it, but then we

will build a separate class that's just

on this subject as part of our SDR

signal intelligence um uh program where

no we are

next year we also have two courses on

mobile hacking with SDR. So, next year

is going to be a big year in new devices

that we'll be putting through. And you

and I did a video on the uh on how the

Mexican cartels built their own private

cell system. Well, we'll be doing that

next year as well. So, we'll be building

showing you or we'll building in class a

private cell system, right? So, look for

that. That's probably about mid

>> 2026.

um somewhere around there early late

spring summer of 2026 we'll squeeze it

in there. So yeah, we got we got a lot

of stuff going on in that field in at

hackers. This is one of them. This is

one of those things that's that's

definitely a signals intelligence

application and I'm pretty excited about

it. I think it's it has a lot of

applications, some good and some bad,

but but you will know how they work and

you will know how to build one.

>> I agree. And this is cool. And I see the

how the community is growing around SDR.

I see much much more people are really

interested in uh that asking a lot of

question. So it's great opportunity for

everyone. Derise are you know spending

their courses and knowledge to to

everyone. It shows out the technology uh

moved forward so much and you know it

raised also many questions like privacy

like

um privacy spy spy you know spyware or

surveillance I will say and um

definitely very insightful

>> you know this came out a couple of years

ago you know and so it's it's been out

there But I don't think people have

really like clicked on it like, "Oh,

this could change so many things. This

is this is like a gamechanging

technology for spies, right? For whoever

whoever's going to use it, it's going to

be gamechanging." So, and I'd be

interested to see how how what distance

we could extend it to with more high

powered higher frequency signals. So,

that most of the SDRs have a limit of

about six gigahertz, right around there.

So, that's about as much as we'd be able

to press it up and and see what we can

do uh at that level versus 2.5 GHz.

>> With that said, OTW, thank you so much.

very insightful, very eyeopening,

definitely much to learn here. And if

you love such content, please leave your

comment in the descriptions. There are

many more videos OTW and I are planning.

So stay tuned, stay uh keep asking me uh

where when is the next video, when OTW

is coming to to the show and OTW as

always, thank you so much.

>> Thank you for having me, Yian. I really

enjoy being on new