Visions of the Future (2007–…): Season 1, Episode 1 - The Intelligence Revolution - full transcript

Three centuries ago,
the great English scientist
Sir Isaac Newton wrote:

"I seem to have been only like a boy
playing on the sea shore

"whilst the great ocean of truth
lay all undiscovered before me."

Today, once again, we are like
children playing on the sea shore,

but the great ocean of truth
is no longer undiscovered.

We have unlocked the secrets
of matter - the atom.

We have unravelled the molecule
of life - DNA.

And we have created a form
of artificial intelligence -
the computer.

The discovery of the fundamental
laws of nature in the 20th century

will open up unparalleled
possibilities for the 21st.

We are making
the historic transition



from the age of scientific discovery
to the age of scientific mastery

in which we will be able
to manipulate and mould nature
almost to our wishes.

I will show you just how rapidly
and radically
science is transforming our lives.

I believe this will confront us with
profound challenges and choices,

so profound that we have to start
to address them today.

In the coming decades,
we'll see how we are becoming
masters of intelligence,

how science will allow us to create
and manipulate intelligence
almost at will.

This mastery of intelligence will
enable us to design new worlds,

but this is just the first step.

From mastering machine intelligence,
we'll move to creating
intelligent machines.

Welcome, Dr Kaku.
Glad to be here, Asimo.

And, ultimately,
we will redesign our own minds.

Driving all this is the exponential
growth of computing power.

This is the old IBM 1401.



Back in the 1960s,
it was state of the art.

It was an engineering marvel.
It filled up an entire room,
it weighed four tons,

and in today's money
it would cost about ?1 million.

The old 1401 could perform just over
4,000 calculations per second.

This is a mobile phone of today.

It costs about ?50,

it has a microchip
the size of your fingernail,

and it can do about one billion
calculations per second.

That's 300,000 times faster
than the old 1401.

The exponential growth of computer
power will profoundly reshape
all of human civilisation.

By 2020, a chip
with today's processing power
will cost about a penny.

That's the cost of scrap paper
that we simply throw away.

And children will wonder
how could we possibly have lived
in a world that was dumb?

Our world is already much smarter
than 10 years ago.

And as computing power doubles
every 18 months,

it's propelling us toward
a very different future.

I'm sitting in this car,
but I'm not steering.

My foot is not on the accelerator,
I'm not driving at all.
The car is driving itself.

This car uses GPS systems
and laser sensors

to find its way and avoid obstacles,
completely autonomously.

And all this achieved
with the processing power
of just eight desktop computers.

Very soon, not just our cars,
but our roads will also be smart.

Built-in microchips will manage
traffic and what once seemed
a futuristic fantasy

will soon become reality.

The city of tomorrow!

Vehicles, electronically paced,

travel routes remarkably safe, swift
and efficient.

I believe that in the future
the very words "traffic accident",
"traffic jam",

will be something that disappears
from the English language.

By 2020,
intelligence will be everywhere.

Not just in a car or the roads,
but practically in every object.

Scientists call it
ubiquitous computing.

What's ahead is pretty inevitable if
you look at the last three decades.

The '80s,
the decade of the microprocessor,

symbolised by the personal computer.
That's when we built our computers.

The '90s, the decade
of networks and communications,
symbolised by the World Wide Web,

was when we connected
our computers together.

was when we connected
our computers together.

In this decade, it's computing that's
everywhere, connected to everything,

and embedded in everything. You'll
know ubiquitous computing has arrived

when you realise
computers have become invisible.

Already we have computers
in our phones, TV sets, stereos,
wristwatches and dishwashers,

but in the future
computers will move even deeper
into the very fabric of our lives.

In just a few years,
basic microchips will be so cheap

they could be built into every
product we buy - an invisible,
intelligent network

hidden in our walls, our furniture,
even our clothing.

Wearable computers will continuously
monitor our health. It's like having
a personalised doctor on call,

24 hours a day.

In case of an emergency, like
a heart attack, your clothing will
upload your entire medical history,

locate your position
and call for an ambulance,
all while you are unconscious.

We will be reliant on computers
and networks to a degree
we've never imagined.

We'll all be a little like Blanche
DuBois in Streetcar Named Desire,

where she says, "I have always relied
upon the kindness of strangers."

In this case, the strangers
will be machines toiling away
in complete anonymity.

You won't even know they exist.

You see this aspirin pill?
Looks pretty ordinary.

But, believe it or not,
in the future, this could have
the power of a PC and a video camera

and it could photograph
all your internal organs.

It's sort of like swallowing
an entire medical laboratory.

So, in the future,
computers could monitor your health
from the outside and the inside.

'But perhaps the biggest impact will
come when ubiquitous intelligence
combines with another technology -

'the internet.'

In the future,
the internet will be everywhere,
including your sunglasses.

It will be projected
from the glasses into your retina.

Or you will use the lens
of your glasses as a movie screen.

In other words,
your eyeglasses could be your
future home entertainment centre.

Your glasses will provide you
with a constant flow of information.

They'll recognise people's faces
and download
their personal profiles.

500 years ago,
the mass distribution of books
set off an intellectual explosion

with the coming of the Renaissance,

but today, with the internet
and ubiquitous computing,
it's an even greater revolution.

First, it's global. It's not just
for the elites, it's for everyone.

Second, it's interactive.
Ideas get dissected.

This is democracy in action.

This revolution is just
the first stage in our mastery
of machine intelligence.

'And it's changing not just
our ideas, but our relationships
with each other.'

So here I am,
dancing in a virtual world.

My dancing partners are real,

but they're located several hundred
miles from here, so I'm dancing with
a virtual three-dimensional image,

transmitted to me in real time.

Simultaneously, my image
is being transmitted to them.

'This new technology is called
tele-immersion.'

Three-dimensional technologies
like tele-immersion have
virtually limitless possibilities.

They could be the evolutionary
successor of the telephone
and might even reduce air travel.

Now, at the present time, the
technology is in its early stages,

but imagine in the future
having an entire business conference
in a virtual world.

Imagine celebrating a virtual
Christmas party with friends and
relatives in a virtual living room,

even if thousands of miles away.

Imagine a virtual television
programme, as if you are sitting
right there in the audience

or right there,
running on the playing field
of an exciting football game.

'And what makes this a credible
scenario is the unprecedented growth
we are already seeing

'in online communication.

'Facebook, MySpace
and instant messaging are moving
into the mainstream.

'And with virtual reality,
it goes one step further.

'We can create virtual worlds
and virtual selves -

'avatars who are not limited
by real-world constraints.'

Online games like Second Life
allow us to create worlds

where everything and anything
is possible. These are worlds
of endless self-reinvention,

where we can change our shape,

our sex,
and even our species.

'As for my avatar,

'I decided to lose a few years.

'This is not just escapist
entertainment - people actually live
in these worlds.

'They even fall in love.'

I chose a plot of land
near a nice-looking build
and it turned out it was his.

Chris had built something
very unconventional, very curvy,
very futuristic. That appealed to me.

It also happened to be in a lovely
setting. A nice, green area
with streams running through it.

I'd scripted a brain to control the
house, tint windows, lock the door,

even make a swimming pool appear
from underground when I wanted it.

Elaine used to pop over
on the pretence
to say hi to Brain.

The first time we ever went out,
we did have a date,

it was on a secluded, romantic island
in Second Life. Yeah.
We hung out and chatted

and acted silly.
Six months after their first meeting
in the virtual world,

Alayne Wartell and Chris Edwards
got married. For real.

Second Life has changed my life
in a lot of ways.

I live in England now
instead of the United States,
I met my husband,

and I have a new career.

Alayne earns real money
selling virtual products
in a virtual world.

'My main business is accessories.

'I make jewellery and shoes
and handbags and things for avatars.

'How you look is a really important
part of being there,

'so I do make a living.'

In fact, I make more than I ever did
in my previous career.

That's amazing to me, that it's come
that far. It's really wonderful.

'I'm surprised how quickly
I'm getting drawn into this world.

'In just four years, Second Life
has grown from a few thousand
enthusiasts to more than 5 million.

'By 2020, there will be
an entire three-dimensional universe
in cyberspace

'with virtual countries
and governments, schools and
universities, property and stocks

'and virtual families and friends.'

Virtual reality is going to become
more and more like real reality,
but have the advantage

that I can share a virtual reality
environment with someone else
hundreds of miles apart.

We can be other people
and change environments quickly.

It has a lot of advantages
over real reality.

I think in 10 years
things like Second Life will be
as prevalent as email is now.

And virtual worlds will become
a similar way for people
to get together, collaborate.

I can see in the future
that it's going to be so much more
capable than it is today.

And I'm going to love it.

When I was a child,
we read about a fictional character
called Walter Mitty.

By daytime, he was a typical
middle class boring individual,

but at night he had very fierce
fantasies living out wild dreams.

In some sense, all of us are
Walter Mittys, but with Second Life
and virtual reality,

we can live out our fantasies.

And in the future, the lines
between our virtual fantasies
and the physical reality

will increasingly blur.

One notion is that virtual reality
interfaces might be integrated
into the human body.

We can have a display built in
to layers within the eye
or into the optic tract somewhere

or indeed into the brain itself.

But these possibilities raise
some rather disturbing questions.

What happens if we assume so many
different identities that we begin
to lose our own sense of identity?

What happens if we begin to prefer
virtual social networks
over our real social networks?

And the family - will the family
suffer if we spend more time
with our virtual family

than our real one?

If you have more than one avatar,
more than one personality,

it could impact very much
on how you see yourself.

Human beings have always had
multiple personalities.

I'm different as a daughter
to as a sister, to as a friend...

We all adapt to the context of
the time and the place that we are,
with the group of people that we are.

Different people bring out different
things in you. But it may be in these
avatar worlds it's magnified tenfold.

What can I do for you?

Already more than 30 million people
worldwide spend an average of
20 hours per week in virtual worlds.

Some, like Sarah Rogers, actually
prefer virtual communication to
socialising in the physical world.

In real life, I'm quite a shy person,
but if you asked
any of my online friends

if I was shy,
they would say absolutely not.

I am able to talk freely,

just because I have the time
to think about what I'm saying.

It's quite liberating
because I feel like I'm more myself
when I'm not under social pressures.

In real life,
Sarah Rogers is a history student.

In the multi-player online game
World of Warcraft
she is an undead priest.

The online games
do compete for my time.

Some days it can be
sort of six hours, eight hours.

However, it's always more exciting
to go and slay some monster

than read about history.

You can be a dragon slayer
or the princess or whatever,

but you don't have body language,
you don't have pheromones,
inner thoughts.

It might be in the future we go,
"Eugh! Real life conversation?

"I couldn't do that! I'd much rather
have this sanitised interaction."

It might be, in the end, a little
bit like that that we become used
to not having real conversations.

There's some evidence of that.

You don't see people face to face

so you can tell them anything
and you don't feel embarrassed.

In a way, you actually get closer to
people than you would in real life
and you form stronger bonds.

Just as the advent of language gave
us an ability to conceive the world

with a vast array of references
and precision,

the ability to create and share
simulated content in virtual worlds

is going to help people conceive
of a multitude of aspects of reality

that we just don't
really consider today.

Virtual reality and ubiquitous
computing are only the first stage

of our mastery
of artificial intelligence.

As we move further into the 21st
century, artificial intelligence
will revolutionise our lives

even more radically

It will result in an evolutionary
leap that will profoundly challenge
the human condition.

In the latter half
of the 21st century, we will enter
a whole new realm of mastery.

We will move from being creators
of machine intelligence to being
creators of intelligent machines,

machines that begin
to rival human intelligence
or perhaps even exceed it.

And this could mean
that a long-held futuristic dream
may eventually come true -

machines that could fulfil
our every whim.

Already large parts of our lives,
our society, our economy

are run by machines with
specialised artificial intelligence.

There are already hundreds of
examples of what I call narrow AI.

Basically, programs
that are operating at human levels,

doing what humans used to do
and doing them, generally, better,
for specific task.

Flying and landing aeroplanes,
guiding weapons, making billions
of dollars of financial decisions.

I could give hundreds of examples.
This used to require humans.

Machines now do it faster,
more accurately. And the narrowness
is gradually getting less narrow.

And, due to the explosion
of computing power,

more and more machines are now being
designed to think for themselves.

Like these modular robots,
called superbots.

Every module, if needed, can become
a brain to control others.

And they can detect
where they are in the body.

Basically, they use their infrared
and they have six-way communication
with their neighbours.

They would say, "Do I have
a neighbour on my right-hand side?
On my left-hand side?"

If it detects it's in the arm,
it will perform like an arm.

If it detects it's in the leg,
it will perform as a leg.

These are just prototypes.

The next generation of superbots
will be able to assess
their environment independently

and decide for themselves
which shape would be best.

It will change its shape to become
a snake in order to go through
a very narrow place,

or change the shape to have legs
to grab things and climb down
certain things.

Eventually they'll perform different
tasks in different environments.

That's our grand goal
for this project.

Up to now, robots' ability to think
and act autonomously has been
limited by two major hurdles.

The first is pattern recognition.

Robots can see - better than us -
but they don't understand
what they are seeing.

They can't recognise objects very
well. The second is more important.

Robots can hear - much better than
us - but they don't necessarily
understand what they are hearing.

Yet this may be about to change.

'Here at the Massachusetts Institute
of Technology, researchers have
recently made a breakthrough.

'They have created a machine
with object recognition skills,

'by combining computer science
with neuroscience.'

Our work has been motivated mainly
by trying to understand
the neuroscience of the brain,

how the brain works.

If you understand
the part of intelligence works,

you can recreate it in machines.

The MIT team designed a computer
that emulates the way the brain
processes visual information.

The machine was then shown hundreds
of sample images of common objects,
like buildings, vehicles,

animals or plants.

Afterwards, the machine was shown
images of street scenes
that contained those objects.

The question was whether
it would recognise them.

The results were astounding.
The computer succeeded in finding
78% of the objects.

It did surprisingly well,
I should say.

Much better
than I would have thought.

Intelligence is the ability
to learn to recognise objects
in difficult situations.

I'm excited about this discovery.

But how does the computer compare
to human abilities?

This computer program and I
are going head-to-head in a contest.

It's going to be man versus machine.

Both of us are going to be
flashed a picture
so fast you can barely see it

and we have to decide whether or not
there's an animal in that picture -
Immediate Object Recognition.

Most computers do horribly at this.

'A picture is flashed up onscreen
for just a fiftieth of a second.

'I then have to decide whether
there is or isn't an animal present.
The computer does the same.

'And then the picture is revealed,
along with our answers.

'Let's see who comes out on top.

'I get the impression that the
computer has got the better of me.

'And, in fact, it has.'

It's rather ingenious. This computer
can mimic the way the human brain
processes information.

And look - it beat me.

'Research into artificial
intelligence is at a turning point.

'In the coming decades,
machines will gradually approach
human level intelligence.

'What kind of relationship will we
develop with intelligent machines?

'To find out, I've travelled
to the country my family came from -
Japan.'

My parents grew up in the Shinto
and Buddhist tradition.

In Shintoism, there are gods and
spirits which inhabit all things,

even inanimate things
like machines and even robots.

And in Buddhism,
the emphasis is on harmony,
not conflict, not domination,

but living in harmony
with the world.

'So the Japanese people conceive
robots as friends, companions
and even confidants.'

Welcome, Dr Kaku.
Glad to meet you, Asimo.

I'll show you to the table.
My pleasure.

'It's no surprise that the most
advanced humanoid robot, Asimo,
has been developed here

'in Japan.'

Thank you for making time
to visit us. No problem.

I'll bring you something to drink.

'While Asimo can recognise obstacles
and pre-registered faces,

'most of what it says and does
is pre-programmed.'
Please take a seat. OK, I will.

Please excuse me.

'And yet Asimo has one
very important skill -
it can walk and move like a human.'

I start walking.

We take it for granted that we can
walk. We don't even think about it.

In fact,
it's an extremely complex task.

It took the engineers at Honda
20 years of research

to achieve Asimo's
human-like movements.

Asimo is one of the most advanced
robots in the world

when it comes to walking, when
it comes to running and mobility.

Things that were once considered
impossible, Asimo can do.

So Asimo is an engineering marvel.
Five or ten years ago, a robot that
could walk with this sophistication

was beyond reach.

'Because Asimo's looks and moves are
so strikingly similar to a human,

'I don't see in it just a machine.'

I came for a delivery. Thank you.
I brought you orange juice.

'I know Asimo is a machine...'

Thank you so much.
'..but I find myself relating to it
as though it was a real person.

'We tend to anthropomorphise things
closest to us.'

For example, recently my wife and I
bought a Roomba,

one of these robot vacuum cleaners
that cleans your floor
automatically.

My wife's attitude toward the
machine began to change with time.

All of a sudden, she began
to call it an affectionate name
and say, "Don't work it so hard.

"Let the poor thing rest.
Give it a break,"
as if it were a real person,

some kind of pet or even a baby.

'The more lifelike the machine, the
more we develop an emotional bond.

'And the more we will tend
to interact with it.

'Yet what will happen when,
increasingly,
the machines interact with us?'

MUSIC: "Surfin' USA"

'These children are visiting
the world's largest robot museum
in the Japanese city of Nagoya.

'Some of the robots on display here
are specially designed to foster
an emotional bond with humans.'

This is the Sony Aibo.
In other words, robo-pet.

The Sony Aibo can register
about six different emotions,

like hunger, distress, pain.

When you want to pet the dog,
it registers pleasure. Pet its back,

tickle its ears or its chin.

And when it runs out of electricity,
it shows that it's hungry.

The children here clearly love toys
like Aibo.

Their reaction gives an insight
into our future relationship
with robots.

I showed some Australian children
of a robot called Curio,
which means curiosity.

This Curio is amazing.

Not only can it walk up and down
stairs, sit down, right itself
and so on,

but it can process conversations
and respond as if it has emotions.

When I asked the eight-year-old
children, "Would you rather have
your best friend

"or have Curio as your best friend?"
the best friend was dumped
immediately in favour of Curio!

APPLAUSE

Some of your closest friends
might not be humans at all.

You may become emotionally attached
to a pet robot or to an artificial
intelligence out on the web.

You'll know that it's not really
intelligent, but you've gotten a
habit of sharing your secrets to it

and it listens so well
and so much better than your friend
that you'll be dependent on it.

'But what if these robots did not
just act as though they had
emotions?

'What if they did really feel pain
and pleasure, sadness and joy?'

Nice or nasty?

Nasty. Because you were meant
to register pain.

I'm not going to hurt you.

Nasty or nice?

Nice.

'Traditionally,
emotions have been seen
as the antithesis of intelligence.

'But scientists are beginning
to understand that emotions are
critical for higher intelligence.'

Scientists have been hesitant
to want to work on emotion.

Scientists think of themselves
as very rational, unemotional beings

and yet what scientists often forget
is that we are guided
in what we choose to focus on

by the mechanisms of emotion.

So we came along about a decade ago
and started saying, "This matters."

We have to understand how emotion
influences intelligence
so we may emulate it in machines.

Don't you care?!
Care? No, why should I care?

'Emotional intelligence is the
cutting edge of human intelligence.'

You wouldn't care about someone
in pain? We do not feel pain.

Being jealous, expressing a loving
sentiment, getting the joke.
These are very complex behaviours,

our most intelligent behaviour.

Pardon me for breathing,
which I never do anyway.

Oh, God, I'm so depressed.

Neuroscientists have found something
absolutely essential
about the human brain.

In patients that suffer brain
damage, the front part of the brain,
the thinking part,

cannot communicate well
with the part that governs emotions.

These people are paralysed
by indecision. They can't evaluate
what's important and what's not.

In the future, we want robots
to tell the difference between
right and wrong, good and bad,

important and not important,
and for that emotions are the key.

'So, inevitably, for robots
to be really useful to us

'they will have to be
emotional beings.'

"Let's build robots
with genuine people personalities,"
they said.

So they tried it out with me.

'And these emotional, intelligent
beings will learn and develop,

'becoming more and more independent
of humans.'

I'm not getting you down, am I?

The government here in Tokyo
is considering the world's
first set of laws

to regulate and redefine
the role of robots in society.

This is going to be increasingly
important as robots begin to assume
more and more human characteristics.

You can't predict what will happen

because in a short time the machine
might develop and learn
and invent and test new ideas

in such variety
that the next day you wouldn't
understand what it's doing.

And it would be too busy to bother
trying to explain it to you.

'As machines evolve both their
intelligence and their ability
to navigate our world,

'they may outgrow human control.'

I think there's a very good chance
machines will be smarter than us.

We're going to lose this brain race.

There are two scenarios here -

the optimistic scenario is that these
new superhuman machines are very
gentle and treat us like pets

and the pessimistic scenario is
that they're not very gentle
and they treat us like food.

Me, I hope for the first scenario.

They're coming towards us.

What...what is it?
I don't know.

'Robots developing their own agenda
and turning against humans is
a staple of science fiction.'

Whenever Hollywood shows you humans
coming in conflict with AIs,

sides are roughly evenly balanced.

'Hollywood can only show you movies

'where the good guys
have a fighting chance.'

Goliath versus David is a story.
Goliath versus a fruit fly is not.

Without you, what could he do?
There's no limit to what he could
do. He could destroy the Earth.

Some people think it's hopeless,

it's only a matter of time before
machines become smarter than us
and put us in zoos behind bars.

I don't think it's that clear
at all. We'll be able to choose

the level of advancement
of the intelligence of our robots.

Why don't you keep out? Don't talk
to me like that. I created you.

We do have a choice in how we create
artificial intelligence.

You've got to be very sure
either that that mind is never
going to want to self-improve

or to want to do anything
that destroys intelligent life.

Treat that gun as if it's loaded.

You can't beat us, Dr Fleming!
Am I likely to get hurt? Yes!

What we do with people is put them
through lots of experiments and tests
and challenges

before they are promoted to the next
level, to see if they are trustworthy
and resilient and able to function.

Similarly, as machines get very
complex and less predictable, and
harder to anticipate their behaviour,

we will have to put them through
many more tests and trials
to prove that they are trustworthy

in more complex situations.

I believe it's ultimately up to us
what kind of intelligent machines
we'll create.

We will decide what relationship
we'll develop with them.

And we'll have another
even more critical choice.

We'll be able to choose the level
of our own intelligence.

As machines become more intelligent,

we'll be able to enhance our own
cognitive and intellectual
capabilities. Here's the irony -

as machines become more like humans,
humans might become
more like machines.

And that may represent
the highest stage
in a mastery of intelligence.

Merging our minds with machines
may sound like science fiction,
but it's already happening.

At the Cleveland Clinic in Ohio,
neurosurgeons implant electrodes

into the brains of chronically
depressed patients. The technique
is called deep brain stimulation

or DBS and it's normally used
for Parkinson's patients.

Deep brain stimulation
or a "brain pacemaker" involves

implanting a tiny electrical wire

in different parts of the brain.
This emits tiny electrical signals
that regulate abnormal brain activity

and rewires the brain to help
these patients' quality of life.

Diane is one of the first
depression patients to be treated
with a brain implant.

I suffered from depression
for 20 years.

My world had sort of shrunk
to my living room and my bedroom.

It just felt like I was carrying
a ton on my shoulders.

And even the thought of having to do
anything that took energy

was so exhausting to me
that I couldn't do it then.
So...

It was just sort of a very dark...
shrunken world.

Diane was desperate to find a cure
for her condition.

I tried very many medications -

MAOI inhibitors, SSRI medications,
anti-psychotics,

anti-depressives,
electro-convulsive therapy...

And all of those things
just made me withdraw more and more.

See the nose?

Rezai's team uses the latest
in MRI scanning technology

to locate specific areas
of the brain for treatment.

See here Diane's brain scan.

These computerised systems allow you
to navigate different parts
of her brain

to find the target area.

Microscopic robots are used
to guide the implants
through the brain.

These tiny micro-robots allow us
to start from the top of the head

to a point that is 8-10 centimetres
deep in her brain.

The operation would last five hours.

After three hours, the electrodes
were placed into position.

They woke me up from the anaesthesia
so I could respond.

And they turned it on
and all of a sudden it was like,
"How are you feeling?"

How are you feeling?

I'm starting to smile.
I feel happy.

Suddenly I was just grinning from
ear to ear, giggling a little bit

and I said,
"Yeah, I like this! This is OK."

You know, it's how I used to feel
when I would go to work

and...

you know, have good weekends
and do things.

But I just haven't felt like that
for so long that...
It's been how many years?

Well, since...really '88.

For me it was instantaneous

to feel SO good.

You just don't want it to stop.

So let me show you the X-rays.
You can see here these wires

that are implanted in the mood parts
of the brain to improve the mood
and energy and anxiety.

These tunnel from the top of
the head down here, down the neck,
to the chest

and to the pacemaker battery.

After her recovery, the process
of fine-tuning the settings
of Diane's brain pacemaker began.

We'll start on the left side. OK.

'That's a real rollercoaster ride!'

They switch it one way and almost
instantaneously you're crying.

I don't feel good.

Things just seem darker and heavier.

And then they, you know,
turned that off and maybe switch it
to another parameter

and then you're feeling better.

So I did that for four or five hours,
three days in a row.

Mm, I think I feel a little lighter,
a little brighter.

Feel better?

It's definitely better? Yes.
No doubt, huh? Yeah.

Definitely lighter and brighter.
Definitely? Yes.

My face feels so relaxed.
Does it? Yeah.

'On the fourth day, we found
a setting that I just was...
a whole new person.'

It's the difference
between night and day.

In my life.

Brain implants have the potential
to treat a wide range of conditions
in the very near future.

The use of brain pacemakers
will be more and more common,

for patients with severe epilepsy,

for patients
with Tourette's syndrome,

and one of the next frontiers
will be in the realm of patients
with severe acquired brain injury,

such as stroke,
and patients with autism

and Alzheimer's, down the line. It's
going to affect many neurological
and psychiatric conditions.

'But implants could do more
than simply repair the brain.

'Soon microchips could begin
to augment and improve it.'

Some scientists predict that in
the future we'll have a memory chip.

We'll be able to file away
and store memories
and vastly improve our memory.

Maybe one day we'll have a vision
chip which could help
victims of strokes and accidents

and some scientists predict that
we could even have a thinking chip

that could vastly enhance our
cognitive and intellectual skills.

When you go out to the late 2020s,
almost everybody will have

some amount of non-biological
intelligence inside their brains.
It will happen very gradually,

and as it gradually becomes
more sophisticated,

you'll enter the 2040s and
the non-biological machine portion
of our intelligence could be

vastly more powerful than
the biological portion. Biological
intelligence will be trivial then,

and ultimately that is really
where the action is.

We, as a species,

are starting to put our
information-processing technology
inside our bodies.

And our technologies
are becoming more biological.

In next 50 years we'll see robots
with more biological components

and people
with more electronic components.

So where are the people going to be
and where are the robots going to be
in 50 or 100 years?

It's an interesting question.

We are at the dawn of a new era

in which we might, literally,
be able to change our mind
with the push of a button.

The proof of principle
already exists.

A team at the University of Southern
California is working on an implant
to replace the hippocampus,

the part of the brain that is
instrumental for storing memories.

People whose hippocampus has been
damaged in an accident may remember
things from before the accident,

but have trouble forming new
memories. In a groundbreaking move,

scientists hope to correct this
by inserting computer chips

directly linked to the nerves of
the brain. I started 30 years ago.

I was involved in early
microprocessor development.

I'm a physicist by training,

and I never really anticipated
becoming a neuroscientist

or taking these and translating them
into a biological system.

First, the team recorded
the electrical responses
of the hippocampus of rats.

Then they designed a computer chip
which mimicked these responses.

The chip was tested by interfacing
it with the brains of rats.

You can improve the response
of the rats by about 50 or 70%,

so we do see an improvement
in the rats' performance.

The next step will be
to develop the chip for humans.

By replacing the damaged hippocampus
of a human,

the scientists aim to treat
Alzheimer's and epilepsy.

It's very exciting work
and it's very fulfilling

if you think eventually
you might impact people's lives.

Currently, brain implants
are only used to treat
certain medical conditions.

The very idea of having electrodes
sticking out of our heads,

well, that's repulsive to some
people, even if they can enhance
our memory and our intelligence.

But in the future,
we just might get used to it

and this technology could become
totally safe, reliable
and unnoticeable.

But this raises an even deeper
question. Precisely how many
of our natural body parts

can we replace with artificial ones
before we begin to lose our sense
of being human?

We'll replace ourselves

by beings that are like us
in some respect and not in others.

They'll be able to learn
a thousand times faster
and live a thousand times longer

and these little changes will make
such large changes that it's hard
to imagine what will happen.

One of the biggest problems
with the 21st century

is how one deals with a world that
is far from a level playing field

and having some enhanced ability
when others have no drinking water.

Would that lead us into a world
where the colonialism
of the 19th century pales

in significance to the speciation,
the differentiation of people into
techno-haves and techno-have nots.

'All revolutions have
winners and losers.
This one is no exception.

'But I would say the big losers
are the people who say'

they don't want to get involved.
They're the ones who will discover

that being a little bit out of touch
will have unpleasant consequences.

This is a revolution where it's not
a good idea to be a bystander.

In the past, we were just
observers of intelligence.

During the Agricultural Revolution,
the best we could do was domesticate
animals and use their intelligence

to secure our food supply.
During the Industrial Revolution,

we could build adding machines and
automated instruments, but that's
about it. Now for the first time,

in the Information Revolution,
we can begin to become
masters of intelligence.

This mastery will offer us
unparalleled freedom
and opportunities.

It has the potential to enrich
our lives more than anything
we've seen before.

But it also forces us to confront
key social issues,

choices that we have to start
to address today.

Subtitles by Subtext
for Red Bee Media Ltd - 2007

Email us at subtitling@bbc.co.uk