Horizon (1964–…): Season 50, Episode 3 - Defeating the Hackers - full transcript

It's a rather unlikely
group of scientists.

They're experts in codes
and code-breaking...

..leading researchers in the
baffling world of quantum physics.

They may have built the most
advanced computer in the universe.

And together, they're taking on
one common enemy...

..hackers.

The greatest threat today
to the world is the keyboard.

In the past, it may have been
nuclear weapons

or weapons of mass destruction.

Today, we see that same
level of capability being exercised

by lone individuals using
keyboards as opposed to bombs.



Hackers are trying to devise ways
to steal our money,

our identities, our secrets.

The internet is a bad neighbourhood.

How often are ne'er-do-wells
coming by to rattle the door?

In the digital world, they're
rattling the door knobs all the time.

But it's not just criminals.

Recently, the extent of government
eavesdropping has been revealed.

And now, powerful cyber-weapons
are being uncovered.

My mouth was, like, wide open,
going, "Oh, my God.

"Oh, my God. Oh, my God."

In this murky world,
scientists are trying to harness

the laws of physics and mathematics
to protect us from the hackers.

Mat Honan considered himself
to be pretty savvy

when it came to security
and the internet.



But last year, he discovered just
how devious hackers can be.

The first clue that something bad
was happening

came when he tried
to charge his phone.

When I went to plug it in,
the phone had this icon on it,

an iTunes icon and a plug, that's
the same kind of thing that you see

the very first time
you turn on an iPhone.

And so I went to connect it
to my computer and when I opened up

my computer, the screen turned grey
and it asked for a four-digit PIN.

And I knew I didn't have
a four-digit PIN, I hadn't set up

a four-digit PIN.

I grabbed my iPad out of my bag.

And my iPad was also
in this reset state

that wanted a password to proceed

and the password that I knew
should have worked didn't work.

At that point, I knew
that I was being hacked.

That was pretty terrifying.

I didn't know what they were
doing at this point.

I had no idea
what their motivation was.

The whole hack took
less than 45 minutes.

By five o'clock, basically, my entire
digital life was wiped out.

Every device I own, everything
I had had been taken over

and almost all of it
completely deleted.

Just about every picture
I'd ever taken of my daughter,

old emails, emails from people
who were no longer alive even.

All kinds of stuff
that was very precious to me.

Mat thought he was
the victim of a classic hack.

Someone had repeatedly tried
to crack his password

and eventually succeeded.

He went online to write
about what happened

and then unexpectedly,
the hackers got in touch with him.

They saw it, they saw that I had
speculated that they had brute-forced

my password and this hacker
got in touch with me to say,

"No, that's not how we did it."

And at that point, I tried
to strike up a dialogue with them

because I wanted to understand
both how things had happened

and why they had happened.

And I basically made a deal
that I wouldn't press charges

if they told me how it was done.

I was angry. I was scared.
I was concerned.

I was a lot of things like that.

But I also realised pretty quickly
that this was an interesting story

from a journalist's perspective.

For Mat, it wasn't just personal.

He's also a writer
for Wired Magazine.

His hackers had discovered a series
of loopholes in the internet

which taken together,
left him completely unprotected.

It wasn't like they used some
crazy cracking programme

to hack into all my stuff.
They didn't make my password.

They didn't break any encryption.

They didn't do any
of that kind of stuff.

What they did was they socially
engineered all of my accounts.

And social engineering is basically
just a fancy term for a con job.

Basically, you con your way in
to a company's or a person's

security system by making them think

that an attacker is actually
a customer.

The first step was to find a way
of stealing his identity

for one of his many online accounts.

Their way in was a simple phone call
to the online shopping service

Amazon.

They gave Amazon a fake credit card
number and added it to my account.

They hung up. They called Amazon back
and they told them

they were locked out of my account
and gave them the credit card number

they had just added to my account.

Once they did that, they were able to
get a temporary password from Amazon.

It was a simple deception,
but effective.

The hackers now owned
his Amazon account.

They didn't go on a shopping spree.

What they were after were the last
four numbers of his credit card

to pull off the next stage
of the con.

On those recent orders, they could
see the last four digits

of the credit card
that I had used to pay.

At the time, Apple was using
those last four digits

as an identity verification method.

Once they had those,
Apple gave my password reset.

They now owned Mat's
Apple accounts.

Now, they could access pretty much
all of his digital life.

The ultimate prize
was his Twitter account, @mat.

For the hackers, a trophy.

And to keep this prize,
with a few clicks,

they destroyed his digital life.

My computer, my iPhone, my iPad.

They deleted my Google account
so that I couldn't get back in there

and kick them out of the Twitter
account again.

It was an interesting chain.

They went from Amazon to Apple
to Google to Twitter.

These hackers knew
the security flaws of the net

and how to use them,
one after another,

to pull off this con.

And they were just teenagers.

It's just online vandalism.

They thought that this was going
to be funny and they were teenagers,

so they didn't think about the
implications of deleting
everything someone owns

and how much precious data
you may have in your life.

Data's quite precious to people now,

it's valuable
and they didn't really see that.

What happened to Mat
is now rather routine.

Credit card stolen,
social media accounts broken into.

These loopholes are now fixed,

but in the anonymous realm
of the internet,

there will always be ways
to steal someone's identity.

But if you thought the havoc that
a couple of teenagers can wreak

is unsettling, wait till you
see what the big boys can do.

It was probably the most
sophisticated hack in history

and it could have gone
completely unexplained...

but for cyber-security experts
Eric Chien and Liam O Murchu.

Right from the word go, there was
just red flags going up everywhere.

You can really feel it. The hairs
on the back of your neck stand up

if it's something
really, really big.

Their job is to investigate
the viruses that pop up

on your computer.

Most malicious software they see
is pretty run-of-the-mill.

But then along came Stuxnet.

This was probably the biggest puzzle
we'd ever seen.

There was no way we were going
to step away until we understood

what was happening with this
particular piece of malware.

Back in 2010, they had no idea
of the significance

of what had just landed
on their desks.

They were just curious because
Stuxnet contained something rare -

a zero-day exploit.

That's a flaw in the code
that no-one is aware of.

Zero days are extremely uncommon.

For Microsoft Windows, there was
only 12 zero-days in all of 2010.

Four of those 12
were inside of Stuxnet.

It was the most sophisticated code
they had ever seen.

And it was dense. Every bit
of code in there was code

that was doing something.

Much of it was written in
a strange programming language.

What we discovered were
big chunks of code

that we just did not recognise.
We had no idea what it was.

We realised it was code for PLCs,

Programmable Logic Controllers,
which are small computers

that control factory equipment
and things like power plants.

Every time Stuxnet infected
a new computer,

it would start hunting
for one of these Programmable
Logic Controllers.

Then it would fingerprint them.
It had to be the right model,
have certain key magic numbers,

the right peripherals,

or things attached to those PLCs
had to have the right hardware.

Once it found that, it would
copy itself onto the PLCs

and then just sit there for a while.
They'd actually sit there

for almost a month just watching
what was going on.

And it had to observe what
it believed was normal operation

of the targeted plant,
of the targeted facility.

Our first theory was that this was
actually trying to commit espionage.

It was trying to steal
design documents

and some sort of industrial
control facility.

But when they discovered
where Stuxnet was spying,

things took a sinister turn.

Basically, when Stuxnet
infects a machine,

it contacts a server to say,
"Look, I've infected a machine."

And we were able to get access
to the logs on those machines

to find out where the most
infections were and it was in Iran.

And so that gave us a hint
that it was trying to attack

something in Iran.

Iran was suspected to be concealing
a nuclear weapons programme.

Now, Eric and Liam had a clue to
what Stuxnet could be hunting for.

But the final piece
of the puzzle came

when they realised two ID numbers
held huge significance.

And then in November, we got
a tip-off from a guy in Holland

who was an expert
in the communication protocol

between the PLCs and the peripherals
that are attached to it.

He had mentioned, "Hey,
you know these peripherals,

"they all have these magic IDs
associated with them

"and there's a catalogue that you
can go look up, these magic IDs."

It would turn out to be the defining
moment of their investigation.

There was quite a moment.
I mean, Liam was searching online

and I was just standing behind him
watching what was coming up

on the screen and when it first
came, immediately there was...

I felt a rush of blood to my face

because I was like,
"Oh. This is not good."

They realised that
they'd stumbled across something
of global significance.

My mouth literally dropped. People
say that, but it literally dropped.

My mouth was wide open, going, "Oh,
my God. Oh, my God. Oh, my God."

The magic numbers were IDs
for frequency converters,

devices which change
the speed of machinery.

But these were specific models
with a dedicated task -

they spin centrifuges
in nuclear facilities.

I was just like, "Oh, no. This is
it. It's uranium enrichment.

"It's nothing else."

By matching up clues
from the code to data

from the International
Atomic Energy Agency,

they could even narrow it down
to one specific nuclear plant,

A place called Natanz.

Once the network was infected,

Stuxnet's devious attack
was designed to unfold like this.

It would then, basically,
try to attack mechanisms.

One is it would speed up
the centrifuges to 1,410 hertz...

..which would cause those aluminium
tubes inside of the centrifuges

to vibrate uncontrollably
and to shatter apart.

And the other was to lower
the speed to two hertz.

So, you can imagine
a kid's top that you spin

and when it gets really slow,
it begins to wobble and fall over.

As the centrifuges span out
of control, Stuxnet would start

to play back data it had recorded

when everything
was working normally.

It's like you see in the movies
where there's a guy watching

CCTV cameras and they patch
in fake footage,

so that the security guards
don't realise

they're currently robbing the safe.
It's exactly what Stuxnet did,

but sort of in this virtual
computer environment.

But the final trick would come

when the operators tried
to shut down the plant.

When they tried to hit
their big red button

that would send a signal
to those PLCs to tell the system

to shut down gracefully.

But Stuxnet infected those PLCs
and cut off that signal

and basically, allowed the attack
to continue to operate.

And it seems to have worked.

Stuxnet reportedly destroyed
around 1,000 centrifuges,

setting Iran's nuclear programme
back by about two years.

But there's one rather important
question left -

who built Stuxnet?

I guess the realisation for me was,

this is not hackers in their basement
who are doing this.

This is the big guns here
who are doing this.

We don't have, unfortunately,
any evidence that tells us
if it's any particular country.

I would say that
it's pretty clear to us

it's at the level of a nation state

and pretty clear someone
who is not an ally of Iran.

And politically motivated to stop
uranium enrichment in Iran,

so that narrows it down,
pretty much narrows it down.

No-one has officially admitted
to being behind it,

but it's been widely reported
that Stuxnet was built by the US

with help from Israel...

..something that neither
country has denied.

Eric and Liam have managed
to take part and understand

the world's first cyber-weapon.

Stuxnet was definitely
a seminal moment.

It really opened Pandora's box.

Before Stuxnet occurred,
people weren't really

practically thinking about
the existence of cyber warfare,

of malicious programmes being able
to literally blow things up.

Stuxnet opened that door and
every country today is talking

both about offence and defence now

on nation to nation,
state cyber-warfare.

In today's digital world, no-one's
quite sure who is hacking who...

..whether it's criminals, teenagers
or even governments.

But with so much at stake,
it's not surprising that

some of the most inventive minds
in science are trying
to make it secure...

..hoping to stay one step
ahead of the hackers.

This man spends much of his time
trying to understand

the murkier world of the internet.

He's worked with
some of the world's largest

and most secretive organisations,
trying to protect their secrets.

He started off life
as a mathematician

and became fascinated with the world
of codes and code breaking.

We've never actually been at a time
where codes were more important.

Almost everything you do today
uses a code.

Every time you log onto an internet
service like Twitter or Facebook

and send your password, every time
you log into internet banking,

all of that information is
protected using encryption code.

Codes have long fascinated
mathematicians

because they are some
of the most beautiful

and addictive problems
they can wrestle with.

And at the heart of everything
that we do on the web

is one sort of number -

prime numbers.

We're surrounded by them every day.

Numbers like seven

and 13.

What's so special about them is that

they can only be divided
by themselves and by one.

But what makes them
so important to codes

is when you combine two of them.

If you take two prime numbers
and multiply them together,

you get something
called a semiprime.

What's interesting about semiprimes
is that it is really difficult

to calculate the numbers that could
have been multiplied together

to form them to get back
to the original values.

Here's an example.

If you multiply two primes
like 11 and 13,

you get 143. That's the easy bit.

But if you're given 143
and you've got to work out

the two original primes,

that takes a long time
to figure out.

Easy multiplication one way
and hard the other.

This is the key to internet codes.

You can make a big semiprime
very quickly,

but try to calculate the two primes
that it's made of

takes a very long time.

So it's a bit like un-frying an egg.
Easy one way, really hard the other.

And the bigger the number,
the longer it takes.

It takes mere seconds to go one way,
but the other way would take

thousands of computers
millions of years.

It's something we all use every day.

A big semiprime is
used as a code word, a key,

to scramble your credit card details
when you buy something online.

But only you and your bank know
the two original primes

that can unscramble it.

These keys are private and secure

because it would take longer
than the age of the universe

for any hacker to figure them out.

This system of public
and private keys

is known as the RSA algorithm.

So that beautiful piece
of mathematics

has fundamentally changed
the world around us.

Without this technology, without
the ability to look up public keys

and form these connections, internet
banking, social media, stock trading,

all the things we take for granted
online, fundamentally wouldn't work.

Our information would be far too
accessible to any prying neighbour.

It's made the hunt for very,
very large prime numbers

one of the most important
quests in maths.

And here's the current largest...

all 5,000 pages of it.

17.5 million digits.
A very big prime number indeed.

Yet divisible
only by itself and one.

But as prime numbers get bigger,

so do the computers
trying to crack them.

All the time, computers
are gaining in power.

All the time, new mathematical
methods are being discovered.

So far, we've stayed ahead
of the code crackers.

But that could just be
a matter of time.

Codes like RSA are effectively
uncrackable

because however powerful
today's PCs are,

they can only process
one computation at a time.

But now scientists are working
on a new form of computer

that harnesses the most complex
physics in the universe.

The world we are all used to
is a rather reassuring place.

The laws of physics mean
we can know where things are,

how fast they are moving

and predict where
they're going to go.

But as things get smaller,
a lot smaller,

they also get a whole lot weirder

as you enter the world
of quantum mechanics.

Quantum is like trying to see music.

It's like even trying to hear
colour. It's very weird.

It's the world that Erik Lucero
studies every day.

Take a single grain of sand
and in that single grain of sand,

there are billions
and billions of atoms

and what we're interested in
is looking at what happens
with a single atom.

These kinds of scales are
where nature shows itself

in a completely different way

and that is this quantum
mechanical nature.

The laws of quantum physics have
baffled the greatest scientists,

even Einstein.

At the smallest scales,
the idea that we can know exactly

where anything is
starts to break down.

The mathematics that describes
the world of the very small

means things can be in many places
at the same time.

One of the very important features
of quantum mechanics

is this idea of superposition.

Superposition is the idea
that a particle can be both

in one place or another place
at the same time.

We speak about it even in
a binary sense, like zero or one.

It can be both zero and one at the
same time which is a very odd idea.

Superposition means that objects
have no fixed location.

They really are in several places
all at the same time.

Quantum physics may be
mind-bogglingly weird,

but it's starting to be
very useful indeed

and it might be a way for Erik
to crack the world's

most powerful codes.

Here at the University
of Santa Barbara,

Erik has constructed
a machine that operates

within this fantastical world.

He's built one of the world's
most advanced quantum computers.

He's harnessed
this quantum weirdness

to design a computer that has
the potential to become

the ultimate code-cracking machine.

But first, it has to get
very, very cold.

We have a dilution refrigerator
and this base plate right here

is what gets a fraction above
absolute zero -

orders of magnitude
colder than space.

All of this machinery exists just
to cool down the computer chip,

the processor.

So, inside of this
specially-engineered box,

we have a quantum processor,
a solid-state quantum processor.

On this chip, there are four cubits.

The cubits themselves are what
are performing the calculation.

Classical computers use data in
the form of bits,

each a zero or a one.

But quantum bits, called cubits,
use the feature of quantum physics

that means things can be
in two places at once.

It can be a zero and a one
and everything in-between

all at the same time.

This gives it the power to do
many calculations simultaneously.

We mount this quantum processor
onto the base plate here

and we then make all these
electrical connections.

Then we're able to move
the quantum information

all around that chip
and actually extract the answer.

From a scientist's point of view,
it's a very exciting tool

that we can probe nature.

It's so fast that it could be
the kind of computer

that finally cracks
RSA encryption.

To prove it in principle, Erik
used his computer to find

the two prime numbers making up
a small semiprime.

And so it's sort of
at the level of technology

that I would say is maybe
like an Atari.

It's kind of 8-bit technology.

It was a very neat toy problem
and we tried to find,

using a quantum processor,
the factors of 15.

I'll let everyone think
about that for a minute,

but that is probably something that
we all can do, even in grade school.

And it took me seven years
to get my physics PhD to do that

with a quantum processor.

What's remarkable is not the answer,
but the way the computer does it.

The quantum chip considers
every possible solution

all at the same time,
instead of sequentially.

And you're collapsing to this
one answer that will actually be

the answer you're after
which is a huge speed up.

You explore all of these possible
places and possible answers

and you get the one that you want.

And we learn, yes, indeed,
15 = 3 x 5.

Erik's proved that quantum computing
has the potential to smash

the codes that protect the internet.

It blows the doors off
of RSA encryption.

All we need is more and more cubits.
We just need a larger
quantum computer.

Really, all that's left
to do is to scale up

this particular architecture.
It's a big task

and there's a lot of very,
very bright people

that are all working towards that.

I think that what's exciting
is that it really puts

kind of a milestone in the ground
about where things are

and what we need to do next.

You do realise you've broken
the internet now?

Oh, yeah. I'm sorry about that.

For now, at least, the web survives.

But if quantum computing
holds the possibility someday

of breaking the world's
most-secure codes,

it may also provide an even cleverer
way of keeping secrets safe.

Quantum mechanics is funky in a kind
of James Brown kind of way.

Very, very funky.

It's strange and counter-intuitive.

Seth Lloyd runs the

Center for Extreme Quantum
Information Theory at MIT.

It's sometimes hard to appreciate

just how extreme this research
can be.

Quantum computers are particularly
fine for teasing out

the subtle interactions between
atoms and molecules

in elementary particles,

or for simulating what
happens as a black hole collapses.

Or, for that matter,
a recent experiment that we did to

actually implement
a version of time travel.

So, you can use quantum computers
for all kinds of exciting things.

And you can use the laws of quantum
physics to create the ultimate

way of sharing secrets.

Current codes that are used to send
information securely over

the internet are called
public key codes,

and they could be broken
by a quantum computer.

But quantum mechanics also supplies
methods for communicating

securely in a way that's
guaranteed by the laws of physics.

So, these methods go under
the name of quantum cryptography.

It's really a way of telling

if someone is eavesdropping
on your conversations.

In the weird world of the very
small,

things can be in more than
one place as once.

But all that changes at the moment
that you actually look

and measure where something is.

It's known as the 'Observer Effect'.

One of the basic principles
about quantum mechanics is that,

when you look at something,
you change it.

And this simple feature
allows you to communicate in a way

that's provably secure.

But the reason it's useful
is that this theory applies to

a photon of light,

which can be used to carry
a message, a one or a zero.

It means that if you were sending
a quantum message,

you can tell
if someone else is observing it.

If there is an eavesdropper
on the line.

A good way to understand quantum
cryptography is to

think of three people -
Alice, Bob and Eve.

Alice wants to send secret
information to Bob

and Eve wants to listen in -
to eavesdrop.

Alice takes her information,
a string of zeros and ones,

or bits, and encodes them on
photons - particles of light.

Now, the encoding is
done in such a way that Eve,

if she looks at these photons,
will inevitably mess them up.

She'll change them in a way that
Alice and Bob can figure out.

So, after Alice has sent
the photons to Bob,

she and Bob can confer to find out

which photons have been
tampered with.

The photons that haven't been
tampered with, the pristine photons,

now constitute a secret key shared
only by Alice and Bob,

whose security is
guaranteed by the laws of physics.

Alice and Bob now have a
secret code word,

one they know no-one had listened
to, which they and only they know,

and they can use this code word to
send their messages.

This system, using the behaviour
of some of the smallest

particles in the universe,
is already being used.

Quantum cryptography is already used
by folks who want extreme security,

by banks and by agencies whose job
is to protect information.

And, nowadays, there are a number
of companies who build quantum

cryptographic systems
and, for a fee,

you too can communicate in complete

and utter privacy guaranteed
by the laws of quantum mechanics.

But whatever the technology,

all codes ultimately have one very
human vulnerability.

No matter what you do with
quantum cryptography,

or any cryptographic system,
there are always going to be...

They are always going to be
susceptible to attack where

Eve ties up Alice and imitates her,

so when Bob thinks he's
communicating with Alice,

he's actually communicating
with Eve.

So, even if you can't crack a code,
it may be possible to get around it.

To pull off an inside job,

whether by someone leaking or
selling secrets.

Perhaps the greatest vulnerability
for anyone trying to keep

a secret isn't the science...but us.

Out there are scientists thinking
dark, paranoid thoughts,

imagining a future where every
computer

in the universe is infected.

Your phone, your laptop,
your work or bank.

In this nightmarish scenario,
the things that scares people most

is not knowing
who is at the other end.

ACOUSTIC GUITAR MUSIC PLAYS

On the face of it, Patrick Lincoln's
real life is rather peaceful...

even content.

But the world that he spends
his life imagining is one in which

threats lurk around every corner.

If you think of it as a neighbourhood
and asking,

"How often are ne'er-do-wells coming
by to rattle the door?"

Trying the doorknob to see
if they can get into your house.

In the digital world, they are
rattling doorknobs all the time.

And therefore I think it is
appropriate for us

to start to be paranoid about what
devices can we really trust

our personal, private,
corporate information to.

And, in the end, moving into an ultra
paranoid mindset where

I can't trust any one device.

He's a leading researcher in a field
called ultra paranoid computing.

Ultra paranoid computing is taking
a point of view that no one

In the past, we've relied
on the unique quality

of a human fingerprint...

..the unique quality of an iris...

but even these things can be stolen.

Unfortunately, those systems
are subject to theft or copying,

so folks can copy a fingerprint

and make something that fools
a fingerprint reader.

Even making copies of irises,
photographs, in some cases,

can fool iris scanners.

So, those are imperfect ways to try
to authenticate that the user

is who they say they are.

So, Patrick turned to a part
of the body that no-one can steal.

He started exploring
whether he could implant

a password into an unconscious
portion of the mind.

Modern cognitive science has found
portions of the brain

that are able to record sequence
information like muscle memory.

The way you learn to ride a bike or
the way to learn to play

a musical instrument, that allows one
to remember long sequences,

but not necessarily have
conscious access to

details of the inside information
in that sequence.

What is the 13th note
of Beethoven's Symphony?

Even if you can play
the symphony on a violin,

you may need to start
at the beginning in order to have

your muscle memory continue through
to that note and then reveal it.

But how do you get
the password in there?

MUSIC: "Eruption" by Van Halen

Now his dark
imaginings are taking shape.

In this paranoid world,

it's not been easy to find
a way of logging on.

But Daniel Sanchez may have found
an intriguing solution.

We have a guitar interface that's

based off of popular rhythm
videogames that people play.

And, essentially, what this is,
is these keys correspond to the

four different
targets on the screen.

The left hand responds to the order
that the circles are scrolling,

and the right hand responds to the
timing. So, essentially,

what you're doing is you're making
a bi-manually coordinated

interception response to the circles
as they cross through the targets.

In other words, using both hands.

The game looks utterly random...

but buried within it is a pattern...

one that repeats nearly 200 times.

Your conscious mind can't
pick it out

but what this is doing is creating
a unique muscle memory.

What we're doing is,
the sequence is repeating.

We don't tell people
the sequence is repeating

and, as they perform
it over and over again,

they become able to perform
a sequence even though

they don't know that they're
learning it.

So, that's how we're able to sort
of store information in people's

brains without them
knowing it's being stored there.

After 45 minutes, the password is
embedded in your muscle memory,

right here in the basal ganglia, a
deep, unconscious part of the brain.

To prove your identity,

you play along with the same
task as before but, this time,

you're actually playing your
password

in your own signature style.

So, essentially, what someone would
do is sit down at a computer

and start performing it. And what
the computer does

is it takes that data and it will
look at their performance

on the trained sequence versus novel
sequences

they've never performed before.

And you can use that information to
say this participant knows

that particular data, or knows that
particular information,

therefore it's Bob. You would have
to know nothing else about them.

It's simply their performance
and their motor abilities that

can tell you who
they are based on what they know.

It may seem strange,

but this could be how you
log on in a paranoid future.

After this entire protocol is done,
a participant will leave

the lab knowing something
they don't know that they know.

That's the password
and the information that we're able

to store that they can't
divulge to anyone else,

and that's essentially how the
cortical cryptography works.

Right now,
were in the grip of a new arms race.

On one side, the code makers
and scientists,

defenders of our digital lives.

On the other side, the hackers
are becoming ever more devious.

Quantum physics and ultra paranoid
computing are just the latest

place where this battle is
being fought out...

..but it is one that is
constantly shifting.

Noisebridge, San Francisco,
a workshop for hackers...

in the original sense of the idea.

A place for pioneers. People taking
apart technology, improving it,

upgrading it, having fun.

But you don't have to look far

to see how connected everything
has become.

Phones with powerful computers,

cars with satellite navigation,

electronic books, even fridges.

And this world of connected
devices is the latest

battleground for the hackers.

Barnaby Jack has been probing
this world

of connected devices,
looking for weakness.

His aim, to hack these
devices before the hackers do.

I've always been doing research,
so I would look at

devices or software,

and I would try and find ways to
break into that code.

And once I found out a way to
break into the code,

I'd write the software that did it.

Hacking proficiently,
I guess I would say,

so I take the same route that
a normal hacker would take

to find these vulnerabilities
and exploit them.

Like any hacker, Barnaby set
out to find the weak points.

The easiest way to bypass
the security systems.

Everyone has wanted to
jump on the wireless bandwagon.

But by going wireless like this,

a lot of people haven't realised the
security ramifications of doing so.

Everything that has a wireless
capability

can potentially be hacked remotely.

So, I decided to look at software
that runs on these devices

because,
once you compromise those devices,

there's a very immediate and real
world effect.

His target was something
we all rely on every day.

Something you might think had
the ultimate security...

Banks.

Or, more precisely, a certain
form of stand alone cash machine.

I decided to look at ATMs because,
you know, they're full of money.

And I looked online,

and I basically just bought them
directly from the distributor.

I took the software off the ATM

and then I reverse engineered
that software,

and I saw that there was a remote
update mechanism.

This was the undefended
part of the system, the way in.

Typically, it would require...

usernames and passwords to access,
but I found a vulnerability which

let me bypass all the username
and password requirements,

and would let me
remotely access the ATM

and upload my own software
anonymously.

Now, the machine was his to control.

It may sound farfetched,
but here's the proof it worked.

And put my software here,
I'd go here and add a group,

so add San Francisco.

I then go ahead and add an ATM,

so I put the name Barnaby's ATM.

So, now I can go ahead and upload
my own software to that ATM.

It connects to the ATM, it sends the
authentication bypass, it succeeds.

And now I could dispense
money from the cassettes,

I could capture people's
credit card details,

I could do all that remotely.

So the software is now uploaded,

so we could go ahead
and issue a remote jackpot command.

That way, anyone near the ATM
at the time can get some money.

So, that could be carried out over
the phone line or over

the...network, whatever it may be.

This flaw, which only affected some
of these sorts of stand alone ATMs,

has, needless to say,
now been fixed.

And Barnaby hopes he gets to these
flaws before the hackers.

We're hoping,
by actually releasing these details

and actually demonstrating
some of these risks,

that the security of these devices
will actually improve quite a bit.

We're working with these
manufacturers to actually

help them improve their codes.

It's estimated there are now over
a billion Wi-Fi-enabled devices,

and hacking and defending
these gadgets

is just the latest battleground.

But perhaps the greatest danger
we face doesn't come from any

one computer, but from
the giant networks

of interconnected computers that

run the most complex
systems on the planet.

From power grids to banking systems
to transport networks.

Because once someone has
hacked one part of it,

they may have hacked
the whole network.

It began with just one computer.

And now it's spreading
through our networks like wildfire.

Power stations are being targeted,

plunging the nation into darkness.

The transport networks are being
targeted, too.

And now the infection is spreading
all across the world.

But this attack isn't real.

It is a simulation being
run by some of Britain's

top cyber security experts.

Cyber security in the UK is
considered to be a tier one threat

alongside terrorism.

This drill is part of a strategy to
pinpoint weaknesses in a network.

They're attacking it to see if,
and where, it breaks.

If we're going to
defend our networks,

we need to understand what an
attacker might do to us.

You need to understand what
the threats are to you,

and you need to understand what
your own vulnerabilities are that

someone might take advantage of.

And if you don't understand what the
attacker might be able to do to you,

you will not develop the best
defences.

The Cyber Range helps us do that.

Once, the only way to test a network
was for a company to attack it

from the inside.

But today, you can come to this
Cyber Range, Europe's first.

This black box is
a kind of internet firing range.

Engineers can programme these 120
computers to create a perfect

mirror image of the company's
global network.

One rack houses their existing
cyber defences,

the other contains the nastiest
malware on earth.

Then the two banks of computers
go to war.

You can emulate a normal day-to-day
email interconnection that

a company would have and,

at the same time, you can
introduce malware into the system.

You can have a look
at the effect it would have.

You can introduce new
software to the system to see how

effective it is against malware.

You can test new intrusion
detection capabilities,

so you can test how well they perform
against intrusions that you

introduce yourself,
all in a safe environment.

The hope is, that by hacking
yourself, you can find those

security flaws and patch them before
a hacker works out how to get in.

Sean McGurk works to protect
America's complex networks,

like power stations
and water companies from attack.

The greatest threat
today to the world is the keyboard.

In the past, it may have been
nuclear weapons

or weapons of mass destruction.

Today, we see that same
level of capability being

exercised by lone individuals using
keyboards, as opposed to bombs.

They can hack into transportation
networks,

into computer networks,
emergency communications networks,

even air transportation are all
susceptible to hackers today.

It's Sean's job to try and find
the unexpected, surprising

weak points that an attacker could
exploit to get into these networks.

What were look at, as far
as vulnerabilities are concerned,

are really three things - people,
processes and technology.

The technology is great.

The encryption is great -
it's very difficult to break.

It takes a tremendous
amount of computing capability,

but the bottom line is

a person can circumvent any layer
of security simply by their actions.

So, in spite of the complex
and sophisticated technology,

once again, it's the people
who are the weak part.

All it can take is something costing
just a few pounds to

get inside the best
protected network.

Removable media is one of the
largest security challenges that we

face today, simply because it
comes in so many shapes and sizes,

so many different forms,

and people are unfamiliar
with its capabilities.

They believe that it's just used to
store files but, unfortunately,

it can also be used to introduce
malicious code

into a network environment.

It may seem unlikely that such
a simple tactic would be effective,

so it's one that Sean was
asked by the US government to test.

When we took as USB stick that had
a corporate logo on it

and placed it in a public area,

we had been a 70 and 80% assurance
that someone would take that

device and insert it
in the corporate network.

When we did the experiment
with a CD ROM

that had the year and pay

and compensation tables just written
with a Sharpie on the disk,

we had almost a 100% guarantee that
piece of media, that CD,

would be introduced
into a corporate environment.

And it is this tactic,
using a removable media device,

which seems to have launched
the world's

most powerful cyber weapon...

Stuxnet.

In 2010, this sophisticated piece
of malware

struck at a uranium enrichment plant
in Iran,

causing significant damage.

This nuclear facility at Natanz
was in a highly secure environment,

cut off from the internet,

but still vulnerable to someone
bringing in a removable

device into the plant.

Whether it was spies or
unwitting accomplices,

we will probably never know.

The challenge with Stuxnet,
for instance,

was it didn't take advantage or try
to break any of the encryption

or the security boundaries,

because it actually exploited
the natural communications

capability of the network.

So, when you plug devices together,
they want to identify each other,

that's part of this plug and play
technology that we use today.

So, these particular individuals
took advantage of that.

They wrote the code to insert into
a network environment

inside the
security perimeter,

so you were already within
the walls of the keep, if you will.

And then it just used
the natural communications

capability of the network,

and it moved from computer
to computer

until it found specifically what
it was looking for.

But nothing in this world of high
stakes hacking...

is quite as simple as it seems...

..because Stuxnet has escaped.

It has now been found
outside its intended target.

What's interesting about Stuxnet

and how we were able to discover
it to begin with is

that it didn't just target
machines in Iran,

it didn't just target
machines in that Natanz facility,

Stuxnet has the ability to
spread to any machine,

any Windows machine across the world.

It has now infected more
than 100,000 machines.

It was never intended to
get in the wild but, unfortunately,

once it did get into the wild,

it demonstrated a level
of sophistication and capability

that up to that point,
no-one had taken advantage of.

This was truly a digital
Pandora's box.

Once it was opened,
you could not put the lid back on.

Stuxnet is now
out in the public domain.

You can take the modules
which are most effective for you

and actually repurpose them,
or retool them,

and launch them
against a private company,

an individual, potentially,
a host nation.

It just depends upon what
your intent and what your desire is.

It highlights the risks of
developing these sorts of weapons.

That they may indeed become
uncontrollable...

and even be used against the nations
that developed them.

There's nothing new about codes
and trying to keep secrets...

..but the advent of global
digital communications

has created a new battleground...

without borders.

One where teenagers...

nation states...

and organised criminals
go head-to-head as equals.

This murky world is
set to become the defining

battleground of the 21st century.

Subtitles by Red Bee Media Ltd