Futurescape (2013) s01e02 Episode Script
Robot Revolution
[ Indistinct shouting .]
-- Captions by VITAC -- CAPTIONS PAID FOR BY DISCOVERY COMMUNICATIONS [ Beeping .]
[ Shouting continues .]
Ridiculous, right? A robot, voting.
Well, you know, they said the same thing about blacks in 1870 and about women in the early 1900s.
But you wouldn't give a programmable robot the full rights of citizenship, now, would you? I mean, they can't think like us Man: They're not people! they don't feel like us They are fundamentally different from us For now.
But the lines are already blurring.
And if you doubt that, well, ask your dad to turn in his pacemaker or your grandma to live without her artificial hip.
Every day, we're getting closer and closer to the moment when man and machine truly meld and redefine what it means to be human.
[ Shouting continues .]
[ Shouting continues .]
Go to [Bleep.]
you fricking robot piece of crap! What is wrong with this guy? What's he so worked up about? Maybe he's having a bad day or a bad life Or maybe he's just wired this way.
You see, we've always thought that the essence of who we are, what separates us from the machines, is some sort of divine spark.
If we remove that spark, does it launch us towards that fateful day when man and machine meld? Recent breakthroughs have allowed scientists to dig deeper into the human mind, to map out our neural wiring.
And guess what.
In the map of the mind, there is no "X" That marks the spot of the human soul.
Get out of here! No soul, no vote! Kaku: What's the big difference between a man and a machine? Well, machines are predictable.
They're programmed.
And when they break down, their repairs are mechanical.
Humans are so complex that we believe there's something called the mind, free will, a soul.
But what happens if, one day, we discover the complete wiring that illuminates how the mind truly works? This technology could eliminate the last barriers between humans and our mechanical creations.
If you have a belief in the soul, and you say, "It's this, and it's that.
" "Our consciousness is this.
Our mind is this" -- you may actually find that there isn't supporting evidence for it.
It could be that the algorithms or the wiring of the brain follows a pattern that's very simple [ Indistinct shouting .]
and that it's not actually that hard to make a mind.
Woods: The first step towards building a mind, perhaps even one that would want the right to vote, is drafting the blueprint of the human brain.
This epic undertaking has already begun at the university of California, Los Angeles.
Dr.
Arthur toga is heading up a groundbreaking project that will do the unthinkable -- map the connections between neurons that underlay all brain function.
The brain is wired in a way where different regions communicate with other regions.
We have embarked upon a project which we call the human connectome project, which is to make a map of this wiring diagram.
Woods: Ultimately, this map will chart roughly 100 billion neurons in the human brain, each one of which could be connected to 10,000 others.
Toga hopes it will give us a complete picture of the mind, how it works And why sometimes it doesn't.
The human connectome project will be able to detect miswirings of the brain, perhaps identify where epilepsy comes from, identifying the places where Alzheimer's disease takes over the human brain, comparable to using a car's diagnostic computer.
Maybe, one day, we'll live in a world without prisons or mental institutions because all that's needed is to tweak somebody's connectome.
Woods: A full diagram of the brain's connectome could also unlock the secret of the self, that essence that makes us so uniquely human.
And this allows us a processing of information that may encompass any aspect by the individual.
Woods: Using a new, souped-up mri scanner, toga tracks the movement of water molecules that flow along the neural fibers.
That pathway shows us how neurons are positioned and how they're connected.
Dr.
toga: There is the top plane and the bottom plane.
From these, we can compute a tract and how it courses through the brain.
Woods: The connections are color-coded to indicate the direction in which they move.
Dr.
toga: The bluish color that you see, these fibers may connect either to motor systems or sensory systems, moving information out to affect motor movements or sensory information back up from processing.
Woods: The color coding illustrates how any given thought can pass through multiple parts of the brain, darting from neuron to neuron to make abstract connections.
When you embark upon such a monumental task, to study an organ system that's capable of performing a Beethoven concerto or ballet or running the 100-meter dash, those are feats of humanness that just gives you awe-inspiring willingness to keep trying to map it.
Woods: Every one of the brain's billions of neurons has its own complex wiring diagram.
Locked inside our skulls are miles of connections between each of those neurons.
And with every memory, every new experience, those connections change.
We used to think that the brain was pretty much set after childhood, but now we know differently.
The brain continues to learn and change all the time, this quality called plasticity.
When we have new experiences, especially if they're intense or we repeat them a lot, we can burn new pathways.
We can also forget old ones.
Dr.
toga: It is a tremendous opportunity to understand our humanness, if you will.
Developing a complete map of the connectome is part of that quest.
Some people say that the connectome is us.
[ Indistinct shouting .]
Our memories, our dreams, our hopes, our defeats -- all of that encoded in our connectome.
One thing that's fascinating about the connectome is if you could really go in and take a scan and then have that brain turn into software, you could have the analog-world version of you and the digital-world equivalent of you.
Woods: So, what happens when the soul is digitized, put on a silicone chip, and made available for download? Allhoff: I mean, I think one of the big questions is, if we could develop conscious, intelligent robots, would they be human? How would that make us feel about our humanity? Woods: With an understanding of the brain as a complex biological computer, will we begin to engineer it like we do machines? Augmenting our minds with digital parts isn't that far off.
But what will it mean for our species when we become more machine than man? [ Electricity crackles .]
[ Razor buzzing .]
[ Ratcheting .]
[ Hiss .]
[ Heart beating .]
All right, congratulations.
You guys are all done.
[ Chuckles .]
No lollipop for that kid.
Just the ability to learn better and faster.
All it takes is a small incision in the skull and a tiny device shoved into his brain.
But what parent would ever agree to that? Unless, of course, everyone else was doing it and your kid was the slow one in school.
I mean, think today to what length parents will go to give their kids an edge in this world.
And we're not talking about fixing some faulty wiring.
We're talking about brain enhancement.
And why would we want that? Our brains are remarkable.
Miraculous, even.
But they can't do everything unless we give them a little high-tech help.
When children see the movie "The matrix" and they see Neo jacking in an electrode and all of a sudden becoming a kung fu master, the first question they ask is, "How can I get one?" well, this does not yet exist, but it's actually physically possible.
Woods: The key to transforming learning from an organic process to a machine-like downloading of information is a squiggly bit of brain known as the hippocampus.
The hippocampus is the gateway to memories.
Short-term memories are stored right here in the prefrontal cortex.
But eventually they have to be transferred to long-term memories, and that's where the hippocampus comes in.
This part of the brain doesn't store the memories, but it does the appropriate conversion.
Woods: At the university of Southern California, bioengineer Ted berger has already proven that a computer chip can replace or enhance brain function.
Right now, what our prosthesis does is to convert a code that's kind of in the middle of the hippocampus to what would be the output of the hippocampus.
They've been able to take mice and access the electrical signals coursing through the hippocampus and record them.
And then, when they shot the message back into the hippocampus, the mouse remembered the task.
We've found that we can not only restore long-term memories, we can enhance the animal's ability to remember.
You could think about using devices like this to greatly enhance human memory and to shorten the cycle for learning, in terms of downloading huge quantities of memory at a single time.
[ Hiss .]
Woods: Chips that augment our hippocampus would very well help us learn faster.
So, will that make them a must-have for competitive parents? At that point, it could create an arms race in elementary school.
Woman: What is 4 times 3? Kaku: Rumors go out that, "Well, Jones' kid, he's been enhanced, and our Johnny has to compete with this enhanced kid.
" the reality is that, with these kinds of technologies, they do not get distributed to everyone at the same time.
Some people get it first.
Some people get it better.
Uldrich: As a society, we have to really think long and hard about who gets this, if it's just the wealthy, that there are real dangers that they will use it to consolidate their power and their wealth.
Woods: A rising class of the intellectual elite would have an additional edge.
They could Usher in a new era of sensory enhancement that outstrips evolution.
We're just actually checking if everything's spiffy here.
Woods: Miguel nicolelis at Duke university is developing implants that could leave millions of years of natural selection in the dust.
They're called brain-machine interfaces.
Nicolelis: In a few decades, brain-machine interfaces will change the way we communicate, the way we interact with machines, and the way we actually interact with one another, for sure.
Woods: Nicolelis believes that the external tools we use today to extend our senses, like radar or heat vision, may one day be implanted directly inside our brains.
Imagine being fitted with the ability to see in pitch black, no glasses needed.
Nicolelis: Well, basically, what we're seeing here is an animal that is being trained to learn a new sensation.
Woods: This lab rat is looking for water by following its feel of infrared light, something no mammal can naturally sense.
What we did was to have a sensor for infrared light in the head of the animal.
And the output of this sensor, that is electrical currents, goes directly to the somatosensory cortex, to the touch cortex.
So, after a few weeks of training, this animal acquired the ability to feel the infrared light that's present in the environment.
Woods: By wiring an external sensor to an interface attached directly to the rat's brain, nicolelis endowed the rat with a new, unnatural sense, as long as he's hooked up.
It's similar to touch, but it's not precisely the same thing.
Woods: This super-rat is hardwired, but the tech exists to go wireless -- and not just for rodents.
The goal is to restore natural senses lost through accidents or illness in humans or replace those sense with new ones.
Nicolelis: Humans will have implants in their brains when it can be done, of course.
In theory, this could be X-ray or radio waves.
It could be any physical energy that we translate into electrical signals to her brain.
Woods: Night vision for cops.
Radiation sensitivity for lab techs.
Or we could all have built-in GPS, like homing pigeons.
But just like we're getting dependent on GPS in our cars, imagine how hooked we could get on those devices if they're built into our bodies.
Kaku: This trend of importing electronics into the human body is happening now because people demand it.
Realize that, already, cochlear implants are now being used by the thousands around the world to give the gift of hearing.
Where does it stop? There's a constant arms race that evolves for cognitive supremacy, and it doesn't just happen within an individual society.
It would happen across the globe.
I mean, how would American leaders respond to the knowledge that brains were being enhanced in China? Woods: Augmenting our brains with computers will change the definition of the human mind.
But what happens when our minds can control machines through the power of thought alone? Will our sense of self expand far beyond our flesh and bones? [ Monitor beeping .]
Woods: Quite an impressive one-man band.
He's performing surgery without nurses, without any assistance at all, except, of course, for the robotic arms he's controlling with his mind.
Now, when we connect our minds to these devices, or even to a full robot surrogate, we'll be extending our selves beyond the very flesh we were born into.
Today work is under way on technology that will seamlessly integrate mind and machine, not only giving us augmented senses, but allowing us to control mechanical devices with our thoughts.
All we have to do is concentrate [ Electricity crackles .]
and hope the power doesn't go out.
[ Electricity crackles .]
Kaku: From the moment we first sharpened a stick to make a spear, we became dependent on technology as an extension of ourselves.
So, mentally controlling a prosthetic arm, a machine, or even a robot with our mind, it's not antihuman.
It's just a part of our natural evolution.
Woods: Scientists at brown university used this tech to help tetraplegic Cathy hutchinson control a robotic arm through thought alone.
But in his lab at Duke, Dr.
nicolelis is pushing this technology to an entirely new level.
Brain-machine interfaces are basically a paradigm that allows us to link living brain tissue with mechanical, electronic, or even visual devices and use the signals that the brain produces to actually control the movements of these devices and interpret what these devices find in the world.
Woods: The key is developing a two-way signal which more closely mimics the relationship we have with our limbs.
Not only do they move at our mind's command, but they also provide our brains with information about the environment.
If it's successful, there's no limit to how far this technology can go.
We'll have the power to move robots by pure thought.
And when that happens, we'll be able to have surrogates.
Woods: The prototype for a full robot replacement is this -- an exoskeleton which can be controlled with a brain-machine interface.
Nicolelis: When we move forward, the human prototype is going to be much lighter and is going to have hydraulic motors.
One of our dreams in two years is having the first kick of the world cup be given by a former paraplegic, a Brazilian teenager, that would actually walk on the pitch and kick the ball using an exoskeleton controlled by some sort of brain-derived signal.
Woods: To achieve that goal, they've been training a monkey to control a digital avatar with his mind.
Once the monkey is adept at controlling the avatar, he can operate a mechanical exoskeleton, turning his thought into mechanical action.
Well, he has to coordinate the two avatar hands and position them in the center of the objects that appear in front of them as fast as they can.
To acquire the electrical signals from the brain that we need, we have microfilaments that are, you know, thinner than a hair, hundreds of them that you can implant in the brain tissue and actually sample the electrical signals the brain cells produce.
Woods: Over time, the computer begins to recognize different signal patterns as specific movement.
Nicolelis: We could read the signals and get this avatar body to be controlled by a brain.
That avatar could also send signals back to the brain so that the brain could interpret what the avatar was touching.
Woods: If we're going to have mechanical extensions of ourselves, we're going to have to train our minds to respond to them.
Nicolelis: What we are discovering here is that monkeys can do that.
So, this is basically the training that, eventually, we are going to do with a patient until we can see that the brain is relearning how to walk or to move.
Woods: In the not-too-distant future, we could have artificial limbs or accessories that feel like natural parts of our bodies.
Nicolelis: If you accept the notion that even our sense of self can be changed or can be expanded to incorporate tools, the concept of being becomes very adaptable, too.
The sense of self doesn't end at the last layer of ourselves, of our skin.
It ends at the last layer of atoms of the tool that we are controlling by our brain.
Kaku: Surrogates will do jobs with the three d's -- dull, dirty, and dangerous.
They'll go into burning buildings, to the bottom of the ocean, to outer space, and we will control them mentally.
And one of the questions that the law will have to face is, is it property or is it a person? Woods: For better or worse, this technology is going to become an integral part of our lives.
Nicolelis: Technology is part of us.
There is an intimate relationship between what we are and what we can create with our brains.
Smart: We're starting to understand the patterns that make us who we are.
That pattern can actually be replicated in technology.
We're learning how to put them into machines.
And that, I think, makes people more open to this idea that I could be intimately connected with my technology in the future.
Woods: Extending ourselves to include machines will open up worlds of possibilities for humanity.
But what if we engineer a machine that thinks? [ Indistinct shouting .]
[ Beeping .]
[ Computerized voice .]
Thank you for your vote.
Merci de vote.
Gracias por votar.
[ Sighs .]
Being human.
A lot trickier than it looks.
You know, our brains have evolved over hundreds of thousands of years into lean, mean calculating machines.
So a device that could keep up with the average human mind, that's the holy grail of neuroscience.
If we could achieve that and make it the size of our skull, put it in a strong, durable robot body, like it or not, you're looking at a potential replacementFor us.
If we knew the trillions of calculations that our brain was performing every time we walk into a room, recognize objects, and pick things up, we would be paralyzed.
As a consequence, evolution has erased all possible consciousness of these calculations.
But with artificial intelligence, we're learning the hard way that these tasks are extremely difficult.
You know all the physics that's going on in the robot, right? Here, we are doing that computation with neurons.
Dr.
kwabena boahen of Stanford university is closing the gap between man and machine by turning what looks like a circuit board into a replica of the human brain, maybe one that could be implanted into a body.
The brain is the best computing machine that we have on this planet so far, and so by building a brain in silicon, I think we can really step forward our ability to design intelligent systems.
Dr.
boahen: There's two things that are very interesting about the brain.
One is very obvious -- that we can outthink any other creature on the planet.
The second is that it does this on a peanut-butter-and-jelly sandwich.
Woods: Boahen is the lead architect of the neurogrid, a staggering piece of hardware that mimics the brain's activity neuron for neuron.
Dr.
boahen: A neurogrid can model a million neurons in the brain in real time.
It's like, you know, an iPod-sized supercomputer.
Woods: The goal is to make a chip that will simulate the human brain and perhaps, one day, become a machine that thinks.
Neurogrid's engineers challenged themselves to rival a supercomputer's power using 100,000 times less energy.
Dr.
boahen: You look at the brain as a physical device.
We don't know exactly how they are hooked up, but it's a machine.
Woods: The key to neurogrid is that it activates links, or synapses, between neurons.
If they're going to replicate a human brain, they need to replicate the circuitry that makes it so efficient.
What we've done with digital computers is we've pushed the precision, precision, precision, precision.
But it's not helping us solve these kinds of problems that humans are good at.
It's the ability to take a lot of data that's sort of fuzzy and make sense of it.
Woods: Living neurons generate unique electrical signals which are the model for the neurogrid.
This is what we call the blueprint of the chip.
This is actually a tiled array of 65,000 neurons.
And if I zoom in, you can make out regular structures.
These regular structures are actually neurons, individual neurons.
Woods: When voltage is applied, just like biological neurons, these individual silicon neurons fire pulses.
Here, you see, in the middle panel, recordings from an actual neuron.
What you see on the top here is the recording from neurogrid neuron, the silicon neuron, which is modeled to behave like this biological neuron.
Woods: That means the only difference between us and a robot that could think like us is more neural pathways.
In other words, it's just a matter of time.
Dr.
boahen: I like to say biology is becoming a technology.
Eventually, we are going to push those limits that the brain has already breached.
Woods: If robots become self-aware, we'll have to ask ourselves what exactly separates man from machine.
And they may ask the same thing.
When robots gain consciousness, I think they will claim to be human.
As kurzweil said, "You may not believe them at first, but they'll be exponentially persuasive.
" one of the key distinctions between the rights offered to humans and the rights offered to self-aware, intelligent machines may be around the right to reproduce.
Because if you have a digital system that can make copies of itself, it can make any number of copies of itself, and all of these would be persons.
And that really messes up the voting rules, at the very least.
Woods: Once we're able to engineer an artificial brain and place it into a human body, it may be hard to distinguish between man and machine.
But building a robot surrogate will require more than re-creating our intellect.
We'll have to teach them how to feel.
[ Indistinct shouting .]
Hello.
[ Indistinct conversation .]
Woods: This human is not just pro-robot.
In fact, she's part robot herself.
Like most of the humans here, she's been modified with a chip that allows her mind to function faster and more efficiently.
And maybe that's why she feels sympathy for a machine.
But what's really surprising here is that our citizen robot feels likewise touched.
[ Indistinct shouting .]
Get a life.
What is a brain without emotion? Would it be a human mind without it? Emotion is a cornerstone of what it means to be a member of our species.
And science has discovered a surprising gateway to human feeling.
When you look at the human brain and you realize that a huge chunk of our cerebral cortex is devoted to our fingertips, you realize that the sense of touch is essential to define who we are.
Touch is so important in human development that people who don't touch or hesitating to touch, that oftentimes, they feel not only physically disconnected from other people, but psychologically and socially disconnected, as well.
So if we want to have robots that have a social and emotional component to them, it's through the intimacy and physical contact that bring us to life.
Woods: Robots that smile, feel relief, or shed tears -- this defies our very definition of robot.
Yet, in the future, that is precisely what our humanoids will do.
And it will all start with the sense that most connects us to one another -- touch.
If robots are going to interact with humans without injuring them, they are going to have to have some of the sensitivity that human caregivers have.
Woods: Today in Southern California, Gerald loeb and his company, syntouch, are developing the tech that could make this type of sensitive robot a reality.
They've already built a piece of it -- a fingerlike sensor called biotac.
The biotac sensor is a tactile sensor that replaces most of what you have in normal biological fingertips.
What we call the bone is epoxy, a nice, hard material that's easily molded around all of the electronics which we put in the bone for safekeeping.
A silicone skin slides around the core and is inflated with a fluid to give it a squishiness very similar to the human fingertip.
Woods: Poke your finger, and you can see an indent from the pressure.
Biotac does the same thing.
Except here, the squishiness helps translate physical contact into analytic data and converts it into human terms.
Loeb: It's hard or soft or it's rough or smooth or it's warm or cold.
Woods: But the real breakthrough design is one that mimics a unique detail in nature's schematics for the human finger.
The skins we initially made were smooth on the outside, 'cause it's a whole lot easier to make smooth skins.
And we decided to try machining a pattern of fingerprint-like ridges, same dimensions as fingerprints, into the mold in which we made the skins to see what would happen.
And lo and behold, the amplitude of the vibrations as we rubbed it over surfaces got 20 or 30 times larger.
Woods: Turns out sensing vibrations is key when it comes to defining texture.
The rougher the surface, the greater the vibrations.
Loeb: Could you design a robot that, if you shook hands with it, you couldn't tell whether you were shaking hands with a robot or a human? Now maybe you're getting at the essence of intelligence.
Nardi: So, if we want machines that are very creative and intelligent, they're ultimately going to have the same kinds of messiness.
So, in the end, it would be very funny to sort of think, yes, we're creating this glorious intelligent robot just as smart as humans, and it's still going to be giving us surprises and unpredictable behavior.
Woods: But the one thing that biotac can't sense, at least not yet, is pain.
Man: If you look, it's black from just picking stuff up.
And there's, like, a burn on there and stuff.
Pain is really an indication of damage.
And in a biological system, you want to prevent further damage by protecting yourself.
Uldrich: Pain is how we learn.
I mean, one of the very first things we learn as children is not to touch that fire.
Kaku: Eventually, robots would become so advanced that they will feel pain and they will really understand the fact that, yes, humans can suffer.
You know, we sometimes ascribe behaviors that look emotional to animals.
You know, if a robot is doing that, who's to say what emotion is? It's really in the eye of the beholder.
Woods: But if we give robots a sense of pain, for their sake and ours, are they still just machines? Kaku: Because it's advantageous to have the ability to feel pain to avoid danger.
At that point, we may begin to say to ourselves, maybe it's wrong.
Maybe it's wrong to make them suffer.
At that point, we may even give them rights.
Woods: Maybe even the right to vote.
But voting, along with many human behaviors, requires a complex understanding of right and wrong.
It's what most distinguishes humans from all other species on the planet.
If we're going to create robots with intelligence and emotion, their ultimate natural evolution will be morality.
[ Computerized voice .]
Hello, Jake.
What would you like to talk about today? Well, my mother-in-law is sick, and my wife can't make it down to visit her.
And I've got a lot of work going on, too, so There's nothing I can do.
[ Beeping .]
Woods: We have robots that can build, clean, calculate, but are we really on the path to robot shrinks? To do that, they need a lot more than logic or even empathy.
They need a deep understanding of the quality that most defines humanity, our moral code of conducts that's called ethics.
And work is already being done to add it to the robot repertoire.
Today when we think of ethical robots, we think, "Who needs that?" but, you see, let's say a robot is guarding the president of the United States, but he sees a bomb, and a child is right next to the bomb.
Does the robot save the child or save the president of the United States? Robots must have a value system.
They must understand what is important, what is precious.
If a robot does not have that ability, we're in deep trouble.
Woods: A husband-and-wife team out of Springfield, Connecticut, have already created a robot that can make his own decisions about what's right or wrong.
[ Computerized voice .]
Hello.
Meet nao, spelled n-a-o, a toddler-sized robot with tactile and visual sensors, vocal ability, and a full range of motion.
He can even tell a good story.
[ Laughing evilly .]
Now, it may look like a cute toy, but thanks to computer scientist Michael Anderson and his wife, philosophy Professor Susan Anderson, nao is also mastering ethics.
Any interaction between a robot and a human has an ethical dimension to it.
When you do incorporate an ethical dimension to these robots, all of a sudden, it opens up horizons you wouldn't see before.
You wouldn't mind having them sit there with your elderly grandfather.
As humans, ethics rule our legal and medical systems and define our moral compass.
Science-fiction author Isaac Asimov famously wrote the three laws of robotics, rules that might one day govern their conduct.
Nardi: The first rule was that robots will not harm another human being, the second rule is that they will obey human beings, and the third rule was that they're going to protect themselves.
Woods: But the Andersons have programmed nao with a different philosophy, one that allows him first to weigh all the options in a given situation and then choose how to act.
This involves being able to recognize ethical features of ethical dilemmas, such as harm, benefit.
All right.
Try.
Woods: To demonstrate this, they give nao a task -- get a prescription from the doctor and administer it to Susan at the appropriate time.
Man: Nao, please take this aspirin to Susan.
But he encounters a dilemma.
Time for your medication, Susan.
No.
No.
Based on what he knows about Susan, nao weighs the pros and the cons of letting her forego the medicine.
Don't look at me like that, nao.
He does this by navigating a series of true/false statements like a flowchart.
Susan: How much harm could be caused if the patient doesn't take the medication? How long it would take for these effects.
Is the patient being unduly influenced by others around him or her? He makes his decision Okay.
I will remind you later.
then follows through with his promise.
Time for your medication, Susan.
You are welcome.
Thank you.
Thank you, nao.
Kaku: Imagine teachers that never make a mistake, police that never take bribes, or judges and politicians that are never swayed by special interests.
Robots with a strong code of ethics could probably teach us a thing or two about being human.
Pretty much anything that we can imagine a human doing, there's a very good chance that over the course of this century, that kind of task could be performed by a robot.
The question then becomes, how does that change how we live? Woods: You could wake up in the morning and not have to go to work, ever.
Would you philosophize? Make great art? Or devolve into a blob? Or would you spend your days hoping and praying the robots don't realize they're subservient slaves and turn on us? It may be that the most important trait we will give robots is mercy, or perhaps they will discover it themselves, along with love.
If we develop purely mechanical systems that emulate, in every possible way, the minds of humans, that essentially have every characteristic of being individual people, do these a.
I.
S, do these machines, have souls? One of the areas that's fascinating is evolving notions of personhood.
Under the law currently, corporations are persons, ships are persons, municipalities are persons.
Why not a transgenically created creature or a cyborg? We will have to determine, if someone creates a part that is incorporated into their body, do they still own that? Do they lease it to you? Or does that become an essential part of you and you own that piece of property? Smart: It's not going to be automatic, adding consciousness to our machines.
But when we do it, it might be the final thing that convinces us that our role is to take the Baton that was given us by the universe and pass it on to our offspring, which are not just biological offspring now.
They're also our machines.
Woods: Long before the day when man and machine become one, we should ask if we're truly passing along the best aspects of ourselves -- the limitless nature of our soul The resiliency of our spirit And the endless possibilities of a mind that has been set free.
-- Captions by VITAC -- CAPTIONS PAID FOR BY DISCOVERY COMMUNICATIONS [ Beeping .]
[ Shouting continues .]
Ridiculous, right? A robot, voting.
Well, you know, they said the same thing about blacks in 1870 and about women in the early 1900s.
But you wouldn't give a programmable robot the full rights of citizenship, now, would you? I mean, they can't think like us Man: They're not people! they don't feel like us They are fundamentally different from us For now.
But the lines are already blurring.
And if you doubt that, well, ask your dad to turn in his pacemaker or your grandma to live without her artificial hip.
Every day, we're getting closer and closer to the moment when man and machine truly meld and redefine what it means to be human.
[ Shouting continues .]
[ Shouting continues .]
Go to [Bleep.]
you fricking robot piece of crap! What is wrong with this guy? What's he so worked up about? Maybe he's having a bad day or a bad life Or maybe he's just wired this way.
You see, we've always thought that the essence of who we are, what separates us from the machines, is some sort of divine spark.
If we remove that spark, does it launch us towards that fateful day when man and machine meld? Recent breakthroughs have allowed scientists to dig deeper into the human mind, to map out our neural wiring.
And guess what.
In the map of the mind, there is no "X" That marks the spot of the human soul.
Get out of here! No soul, no vote! Kaku: What's the big difference between a man and a machine? Well, machines are predictable.
They're programmed.
And when they break down, their repairs are mechanical.
Humans are so complex that we believe there's something called the mind, free will, a soul.
But what happens if, one day, we discover the complete wiring that illuminates how the mind truly works? This technology could eliminate the last barriers between humans and our mechanical creations.
If you have a belief in the soul, and you say, "It's this, and it's that.
" "Our consciousness is this.
Our mind is this" -- you may actually find that there isn't supporting evidence for it.
It could be that the algorithms or the wiring of the brain follows a pattern that's very simple [ Indistinct shouting .]
and that it's not actually that hard to make a mind.
Woods: The first step towards building a mind, perhaps even one that would want the right to vote, is drafting the blueprint of the human brain.
This epic undertaking has already begun at the university of California, Los Angeles.
Dr.
Arthur toga is heading up a groundbreaking project that will do the unthinkable -- map the connections between neurons that underlay all brain function.
The brain is wired in a way where different regions communicate with other regions.
We have embarked upon a project which we call the human connectome project, which is to make a map of this wiring diagram.
Woods: Ultimately, this map will chart roughly 100 billion neurons in the human brain, each one of which could be connected to 10,000 others.
Toga hopes it will give us a complete picture of the mind, how it works And why sometimes it doesn't.
The human connectome project will be able to detect miswirings of the brain, perhaps identify where epilepsy comes from, identifying the places where Alzheimer's disease takes over the human brain, comparable to using a car's diagnostic computer.
Maybe, one day, we'll live in a world without prisons or mental institutions because all that's needed is to tweak somebody's connectome.
Woods: A full diagram of the brain's connectome could also unlock the secret of the self, that essence that makes us so uniquely human.
And this allows us a processing of information that may encompass any aspect by the individual.
Woods: Using a new, souped-up mri scanner, toga tracks the movement of water molecules that flow along the neural fibers.
That pathway shows us how neurons are positioned and how they're connected.
Dr.
toga: There is the top plane and the bottom plane.
From these, we can compute a tract and how it courses through the brain.
Woods: The connections are color-coded to indicate the direction in which they move.
Dr.
toga: The bluish color that you see, these fibers may connect either to motor systems or sensory systems, moving information out to affect motor movements or sensory information back up from processing.
Woods: The color coding illustrates how any given thought can pass through multiple parts of the brain, darting from neuron to neuron to make abstract connections.
When you embark upon such a monumental task, to study an organ system that's capable of performing a Beethoven concerto or ballet or running the 100-meter dash, those are feats of humanness that just gives you awe-inspiring willingness to keep trying to map it.
Woods: Every one of the brain's billions of neurons has its own complex wiring diagram.
Locked inside our skulls are miles of connections between each of those neurons.
And with every memory, every new experience, those connections change.
We used to think that the brain was pretty much set after childhood, but now we know differently.
The brain continues to learn and change all the time, this quality called plasticity.
When we have new experiences, especially if they're intense or we repeat them a lot, we can burn new pathways.
We can also forget old ones.
Dr.
toga: It is a tremendous opportunity to understand our humanness, if you will.
Developing a complete map of the connectome is part of that quest.
Some people say that the connectome is us.
[ Indistinct shouting .]
Our memories, our dreams, our hopes, our defeats -- all of that encoded in our connectome.
One thing that's fascinating about the connectome is if you could really go in and take a scan and then have that brain turn into software, you could have the analog-world version of you and the digital-world equivalent of you.
Woods: So, what happens when the soul is digitized, put on a silicone chip, and made available for download? Allhoff: I mean, I think one of the big questions is, if we could develop conscious, intelligent robots, would they be human? How would that make us feel about our humanity? Woods: With an understanding of the brain as a complex biological computer, will we begin to engineer it like we do machines? Augmenting our minds with digital parts isn't that far off.
But what will it mean for our species when we become more machine than man? [ Electricity crackles .]
[ Razor buzzing .]
[ Ratcheting .]
[ Hiss .]
[ Heart beating .]
All right, congratulations.
You guys are all done.
[ Chuckles .]
No lollipop for that kid.
Just the ability to learn better and faster.
All it takes is a small incision in the skull and a tiny device shoved into his brain.
But what parent would ever agree to that? Unless, of course, everyone else was doing it and your kid was the slow one in school.
I mean, think today to what length parents will go to give their kids an edge in this world.
And we're not talking about fixing some faulty wiring.
We're talking about brain enhancement.
And why would we want that? Our brains are remarkable.
Miraculous, even.
But they can't do everything unless we give them a little high-tech help.
When children see the movie "The matrix" and they see Neo jacking in an electrode and all of a sudden becoming a kung fu master, the first question they ask is, "How can I get one?" well, this does not yet exist, but it's actually physically possible.
Woods: The key to transforming learning from an organic process to a machine-like downloading of information is a squiggly bit of brain known as the hippocampus.
The hippocampus is the gateway to memories.
Short-term memories are stored right here in the prefrontal cortex.
But eventually they have to be transferred to long-term memories, and that's where the hippocampus comes in.
This part of the brain doesn't store the memories, but it does the appropriate conversion.
Woods: At the university of Southern California, bioengineer Ted berger has already proven that a computer chip can replace or enhance brain function.
Right now, what our prosthesis does is to convert a code that's kind of in the middle of the hippocampus to what would be the output of the hippocampus.
They've been able to take mice and access the electrical signals coursing through the hippocampus and record them.
And then, when they shot the message back into the hippocampus, the mouse remembered the task.
We've found that we can not only restore long-term memories, we can enhance the animal's ability to remember.
You could think about using devices like this to greatly enhance human memory and to shorten the cycle for learning, in terms of downloading huge quantities of memory at a single time.
[ Hiss .]
Woods: Chips that augment our hippocampus would very well help us learn faster.
So, will that make them a must-have for competitive parents? At that point, it could create an arms race in elementary school.
Woman: What is 4 times 3? Kaku: Rumors go out that, "Well, Jones' kid, he's been enhanced, and our Johnny has to compete with this enhanced kid.
" the reality is that, with these kinds of technologies, they do not get distributed to everyone at the same time.
Some people get it first.
Some people get it better.
Uldrich: As a society, we have to really think long and hard about who gets this, if it's just the wealthy, that there are real dangers that they will use it to consolidate their power and their wealth.
Woods: A rising class of the intellectual elite would have an additional edge.
They could Usher in a new era of sensory enhancement that outstrips evolution.
We're just actually checking if everything's spiffy here.
Woods: Miguel nicolelis at Duke university is developing implants that could leave millions of years of natural selection in the dust.
They're called brain-machine interfaces.
Nicolelis: In a few decades, brain-machine interfaces will change the way we communicate, the way we interact with machines, and the way we actually interact with one another, for sure.
Woods: Nicolelis believes that the external tools we use today to extend our senses, like radar or heat vision, may one day be implanted directly inside our brains.
Imagine being fitted with the ability to see in pitch black, no glasses needed.
Nicolelis: Well, basically, what we're seeing here is an animal that is being trained to learn a new sensation.
Woods: This lab rat is looking for water by following its feel of infrared light, something no mammal can naturally sense.
What we did was to have a sensor for infrared light in the head of the animal.
And the output of this sensor, that is electrical currents, goes directly to the somatosensory cortex, to the touch cortex.
So, after a few weeks of training, this animal acquired the ability to feel the infrared light that's present in the environment.
Woods: By wiring an external sensor to an interface attached directly to the rat's brain, nicolelis endowed the rat with a new, unnatural sense, as long as he's hooked up.
It's similar to touch, but it's not precisely the same thing.
Woods: This super-rat is hardwired, but the tech exists to go wireless -- and not just for rodents.
The goal is to restore natural senses lost through accidents or illness in humans or replace those sense with new ones.
Nicolelis: Humans will have implants in their brains when it can be done, of course.
In theory, this could be X-ray or radio waves.
It could be any physical energy that we translate into electrical signals to her brain.
Woods: Night vision for cops.
Radiation sensitivity for lab techs.
Or we could all have built-in GPS, like homing pigeons.
But just like we're getting dependent on GPS in our cars, imagine how hooked we could get on those devices if they're built into our bodies.
Kaku: This trend of importing electronics into the human body is happening now because people demand it.
Realize that, already, cochlear implants are now being used by the thousands around the world to give the gift of hearing.
Where does it stop? There's a constant arms race that evolves for cognitive supremacy, and it doesn't just happen within an individual society.
It would happen across the globe.
I mean, how would American leaders respond to the knowledge that brains were being enhanced in China? Woods: Augmenting our brains with computers will change the definition of the human mind.
But what happens when our minds can control machines through the power of thought alone? Will our sense of self expand far beyond our flesh and bones? [ Monitor beeping .]
Woods: Quite an impressive one-man band.
He's performing surgery without nurses, without any assistance at all, except, of course, for the robotic arms he's controlling with his mind.
Now, when we connect our minds to these devices, or even to a full robot surrogate, we'll be extending our selves beyond the very flesh we were born into.
Today work is under way on technology that will seamlessly integrate mind and machine, not only giving us augmented senses, but allowing us to control mechanical devices with our thoughts.
All we have to do is concentrate [ Electricity crackles .]
and hope the power doesn't go out.
[ Electricity crackles .]
Kaku: From the moment we first sharpened a stick to make a spear, we became dependent on technology as an extension of ourselves.
So, mentally controlling a prosthetic arm, a machine, or even a robot with our mind, it's not antihuman.
It's just a part of our natural evolution.
Woods: Scientists at brown university used this tech to help tetraplegic Cathy hutchinson control a robotic arm through thought alone.
But in his lab at Duke, Dr.
nicolelis is pushing this technology to an entirely new level.
Brain-machine interfaces are basically a paradigm that allows us to link living brain tissue with mechanical, electronic, or even visual devices and use the signals that the brain produces to actually control the movements of these devices and interpret what these devices find in the world.
Woods: The key is developing a two-way signal which more closely mimics the relationship we have with our limbs.
Not only do they move at our mind's command, but they also provide our brains with information about the environment.
If it's successful, there's no limit to how far this technology can go.
We'll have the power to move robots by pure thought.
And when that happens, we'll be able to have surrogates.
Woods: The prototype for a full robot replacement is this -- an exoskeleton which can be controlled with a brain-machine interface.
Nicolelis: When we move forward, the human prototype is going to be much lighter and is going to have hydraulic motors.
One of our dreams in two years is having the first kick of the world cup be given by a former paraplegic, a Brazilian teenager, that would actually walk on the pitch and kick the ball using an exoskeleton controlled by some sort of brain-derived signal.
Woods: To achieve that goal, they've been training a monkey to control a digital avatar with his mind.
Once the monkey is adept at controlling the avatar, he can operate a mechanical exoskeleton, turning his thought into mechanical action.
Well, he has to coordinate the two avatar hands and position them in the center of the objects that appear in front of them as fast as they can.
To acquire the electrical signals from the brain that we need, we have microfilaments that are, you know, thinner than a hair, hundreds of them that you can implant in the brain tissue and actually sample the electrical signals the brain cells produce.
Woods: Over time, the computer begins to recognize different signal patterns as specific movement.
Nicolelis: We could read the signals and get this avatar body to be controlled by a brain.
That avatar could also send signals back to the brain so that the brain could interpret what the avatar was touching.
Woods: If we're going to have mechanical extensions of ourselves, we're going to have to train our minds to respond to them.
Nicolelis: What we are discovering here is that monkeys can do that.
So, this is basically the training that, eventually, we are going to do with a patient until we can see that the brain is relearning how to walk or to move.
Woods: In the not-too-distant future, we could have artificial limbs or accessories that feel like natural parts of our bodies.
Nicolelis: If you accept the notion that even our sense of self can be changed or can be expanded to incorporate tools, the concept of being becomes very adaptable, too.
The sense of self doesn't end at the last layer of ourselves, of our skin.
It ends at the last layer of atoms of the tool that we are controlling by our brain.
Kaku: Surrogates will do jobs with the three d's -- dull, dirty, and dangerous.
They'll go into burning buildings, to the bottom of the ocean, to outer space, and we will control them mentally.
And one of the questions that the law will have to face is, is it property or is it a person? Woods: For better or worse, this technology is going to become an integral part of our lives.
Nicolelis: Technology is part of us.
There is an intimate relationship between what we are and what we can create with our brains.
Smart: We're starting to understand the patterns that make us who we are.
That pattern can actually be replicated in technology.
We're learning how to put them into machines.
And that, I think, makes people more open to this idea that I could be intimately connected with my technology in the future.
Woods: Extending ourselves to include machines will open up worlds of possibilities for humanity.
But what if we engineer a machine that thinks? [ Indistinct shouting .]
[ Beeping .]
[ Computerized voice .]
Thank you for your vote.
Merci de vote.
Gracias por votar.
[ Sighs .]
Being human.
A lot trickier than it looks.
You know, our brains have evolved over hundreds of thousands of years into lean, mean calculating machines.
So a device that could keep up with the average human mind, that's the holy grail of neuroscience.
If we could achieve that and make it the size of our skull, put it in a strong, durable robot body, like it or not, you're looking at a potential replacementFor us.
If we knew the trillions of calculations that our brain was performing every time we walk into a room, recognize objects, and pick things up, we would be paralyzed.
As a consequence, evolution has erased all possible consciousness of these calculations.
But with artificial intelligence, we're learning the hard way that these tasks are extremely difficult.
You know all the physics that's going on in the robot, right? Here, we are doing that computation with neurons.
Dr.
kwabena boahen of Stanford university is closing the gap between man and machine by turning what looks like a circuit board into a replica of the human brain, maybe one that could be implanted into a body.
The brain is the best computing machine that we have on this planet so far, and so by building a brain in silicon, I think we can really step forward our ability to design intelligent systems.
Dr.
boahen: There's two things that are very interesting about the brain.
One is very obvious -- that we can outthink any other creature on the planet.
The second is that it does this on a peanut-butter-and-jelly sandwich.
Woods: Boahen is the lead architect of the neurogrid, a staggering piece of hardware that mimics the brain's activity neuron for neuron.
Dr.
boahen: A neurogrid can model a million neurons in the brain in real time.
It's like, you know, an iPod-sized supercomputer.
Woods: The goal is to make a chip that will simulate the human brain and perhaps, one day, become a machine that thinks.
Neurogrid's engineers challenged themselves to rival a supercomputer's power using 100,000 times less energy.
Dr.
boahen: You look at the brain as a physical device.
We don't know exactly how they are hooked up, but it's a machine.
Woods: The key to neurogrid is that it activates links, or synapses, between neurons.
If they're going to replicate a human brain, they need to replicate the circuitry that makes it so efficient.
What we've done with digital computers is we've pushed the precision, precision, precision, precision.
But it's not helping us solve these kinds of problems that humans are good at.
It's the ability to take a lot of data that's sort of fuzzy and make sense of it.
Woods: Living neurons generate unique electrical signals which are the model for the neurogrid.
This is what we call the blueprint of the chip.
This is actually a tiled array of 65,000 neurons.
And if I zoom in, you can make out regular structures.
These regular structures are actually neurons, individual neurons.
Woods: When voltage is applied, just like biological neurons, these individual silicon neurons fire pulses.
Here, you see, in the middle panel, recordings from an actual neuron.
What you see on the top here is the recording from neurogrid neuron, the silicon neuron, which is modeled to behave like this biological neuron.
Woods: That means the only difference between us and a robot that could think like us is more neural pathways.
In other words, it's just a matter of time.
Dr.
boahen: I like to say biology is becoming a technology.
Eventually, we are going to push those limits that the brain has already breached.
Woods: If robots become self-aware, we'll have to ask ourselves what exactly separates man from machine.
And they may ask the same thing.
When robots gain consciousness, I think they will claim to be human.
As kurzweil said, "You may not believe them at first, but they'll be exponentially persuasive.
" one of the key distinctions between the rights offered to humans and the rights offered to self-aware, intelligent machines may be around the right to reproduce.
Because if you have a digital system that can make copies of itself, it can make any number of copies of itself, and all of these would be persons.
And that really messes up the voting rules, at the very least.
Woods: Once we're able to engineer an artificial brain and place it into a human body, it may be hard to distinguish between man and machine.
But building a robot surrogate will require more than re-creating our intellect.
We'll have to teach them how to feel.
[ Indistinct shouting .]
Hello.
[ Indistinct conversation .]
Woods: This human is not just pro-robot.
In fact, she's part robot herself.
Like most of the humans here, she's been modified with a chip that allows her mind to function faster and more efficiently.
And maybe that's why she feels sympathy for a machine.
But what's really surprising here is that our citizen robot feels likewise touched.
[ Indistinct shouting .]
Get a life.
What is a brain without emotion? Would it be a human mind without it? Emotion is a cornerstone of what it means to be a member of our species.
And science has discovered a surprising gateway to human feeling.
When you look at the human brain and you realize that a huge chunk of our cerebral cortex is devoted to our fingertips, you realize that the sense of touch is essential to define who we are.
Touch is so important in human development that people who don't touch or hesitating to touch, that oftentimes, they feel not only physically disconnected from other people, but psychologically and socially disconnected, as well.
So if we want to have robots that have a social and emotional component to them, it's through the intimacy and physical contact that bring us to life.
Woods: Robots that smile, feel relief, or shed tears -- this defies our very definition of robot.
Yet, in the future, that is precisely what our humanoids will do.
And it will all start with the sense that most connects us to one another -- touch.
If robots are going to interact with humans without injuring them, they are going to have to have some of the sensitivity that human caregivers have.
Woods: Today in Southern California, Gerald loeb and his company, syntouch, are developing the tech that could make this type of sensitive robot a reality.
They've already built a piece of it -- a fingerlike sensor called biotac.
The biotac sensor is a tactile sensor that replaces most of what you have in normal biological fingertips.
What we call the bone is epoxy, a nice, hard material that's easily molded around all of the electronics which we put in the bone for safekeeping.
A silicone skin slides around the core and is inflated with a fluid to give it a squishiness very similar to the human fingertip.
Woods: Poke your finger, and you can see an indent from the pressure.
Biotac does the same thing.
Except here, the squishiness helps translate physical contact into analytic data and converts it into human terms.
Loeb: It's hard or soft or it's rough or smooth or it's warm or cold.
Woods: But the real breakthrough design is one that mimics a unique detail in nature's schematics for the human finger.
The skins we initially made were smooth on the outside, 'cause it's a whole lot easier to make smooth skins.
And we decided to try machining a pattern of fingerprint-like ridges, same dimensions as fingerprints, into the mold in which we made the skins to see what would happen.
And lo and behold, the amplitude of the vibrations as we rubbed it over surfaces got 20 or 30 times larger.
Woods: Turns out sensing vibrations is key when it comes to defining texture.
The rougher the surface, the greater the vibrations.
Loeb: Could you design a robot that, if you shook hands with it, you couldn't tell whether you were shaking hands with a robot or a human? Now maybe you're getting at the essence of intelligence.
Nardi: So, if we want machines that are very creative and intelligent, they're ultimately going to have the same kinds of messiness.
So, in the end, it would be very funny to sort of think, yes, we're creating this glorious intelligent robot just as smart as humans, and it's still going to be giving us surprises and unpredictable behavior.
Woods: But the one thing that biotac can't sense, at least not yet, is pain.
Man: If you look, it's black from just picking stuff up.
And there's, like, a burn on there and stuff.
Pain is really an indication of damage.
And in a biological system, you want to prevent further damage by protecting yourself.
Uldrich: Pain is how we learn.
I mean, one of the very first things we learn as children is not to touch that fire.
Kaku: Eventually, robots would become so advanced that they will feel pain and they will really understand the fact that, yes, humans can suffer.
You know, we sometimes ascribe behaviors that look emotional to animals.
You know, if a robot is doing that, who's to say what emotion is? It's really in the eye of the beholder.
Woods: But if we give robots a sense of pain, for their sake and ours, are they still just machines? Kaku: Because it's advantageous to have the ability to feel pain to avoid danger.
At that point, we may begin to say to ourselves, maybe it's wrong.
Maybe it's wrong to make them suffer.
At that point, we may even give them rights.
Woods: Maybe even the right to vote.
But voting, along with many human behaviors, requires a complex understanding of right and wrong.
It's what most distinguishes humans from all other species on the planet.
If we're going to create robots with intelligence and emotion, their ultimate natural evolution will be morality.
[ Computerized voice .]
Hello, Jake.
What would you like to talk about today? Well, my mother-in-law is sick, and my wife can't make it down to visit her.
And I've got a lot of work going on, too, so There's nothing I can do.
[ Beeping .]
Woods: We have robots that can build, clean, calculate, but are we really on the path to robot shrinks? To do that, they need a lot more than logic or even empathy.
They need a deep understanding of the quality that most defines humanity, our moral code of conducts that's called ethics.
And work is already being done to add it to the robot repertoire.
Today when we think of ethical robots, we think, "Who needs that?" but, you see, let's say a robot is guarding the president of the United States, but he sees a bomb, and a child is right next to the bomb.
Does the robot save the child or save the president of the United States? Robots must have a value system.
They must understand what is important, what is precious.
If a robot does not have that ability, we're in deep trouble.
Woods: A husband-and-wife team out of Springfield, Connecticut, have already created a robot that can make his own decisions about what's right or wrong.
[ Computerized voice .]
Hello.
Meet nao, spelled n-a-o, a toddler-sized robot with tactile and visual sensors, vocal ability, and a full range of motion.
He can even tell a good story.
[ Laughing evilly .]
Now, it may look like a cute toy, but thanks to computer scientist Michael Anderson and his wife, philosophy Professor Susan Anderson, nao is also mastering ethics.
Any interaction between a robot and a human has an ethical dimension to it.
When you do incorporate an ethical dimension to these robots, all of a sudden, it opens up horizons you wouldn't see before.
You wouldn't mind having them sit there with your elderly grandfather.
As humans, ethics rule our legal and medical systems and define our moral compass.
Science-fiction author Isaac Asimov famously wrote the three laws of robotics, rules that might one day govern their conduct.
Nardi: The first rule was that robots will not harm another human being, the second rule is that they will obey human beings, and the third rule was that they're going to protect themselves.
Woods: But the Andersons have programmed nao with a different philosophy, one that allows him first to weigh all the options in a given situation and then choose how to act.
This involves being able to recognize ethical features of ethical dilemmas, such as harm, benefit.
All right.
Try.
Woods: To demonstrate this, they give nao a task -- get a prescription from the doctor and administer it to Susan at the appropriate time.
Man: Nao, please take this aspirin to Susan.
But he encounters a dilemma.
Time for your medication, Susan.
No.
No.
Based on what he knows about Susan, nao weighs the pros and the cons of letting her forego the medicine.
Don't look at me like that, nao.
He does this by navigating a series of true/false statements like a flowchart.
Susan: How much harm could be caused if the patient doesn't take the medication? How long it would take for these effects.
Is the patient being unduly influenced by others around him or her? He makes his decision Okay.
I will remind you later.
then follows through with his promise.
Time for your medication, Susan.
You are welcome.
Thank you.
Thank you, nao.
Kaku: Imagine teachers that never make a mistake, police that never take bribes, or judges and politicians that are never swayed by special interests.
Robots with a strong code of ethics could probably teach us a thing or two about being human.
Pretty much anything that we can imagine a human doing, there's a very good chance that over the course of this century, that kind of task could be performed by a robot.
The question then becomes, how does that change how we live? Woods: You could wake up in the morning and not have to go to work, ever.
Would you philosophize? Make great art? Or devolve into a blob? Or would you spend your days hoping and praying the robots don't realize they're subservient slaves and turn on us? It may be that the most important trait we will give robots is mercy, or perhaps they will discover it themselves, along with love.
If we develop purely mechanical systems that emulate, in every possible way, the minds of humans, that essentially have every characteristic of being individual people, do these a.
I.
S, do these machines, have souls? One of the areas that's fascinating is evolving notions of personhood.
Under the law currently, corporations are persons, ships are persons, municipalities are persons.
Why not a transgenically created creature or a cyborg? We will have to determine, if someone creates a part that is incorporated into their body, do they still own that? Do they lease it to you? Or does that become an essential part of you and you own that piece of property? Smart: It's not going to be automatic, adding consciousness to our machines.
But when we do it, it might be the final thing that convinces us that our role is to take the Baton that was given us by the universe and pass it on to our offspring, which are not just biological offspring now.
They're also our machines.
Woods: Long before the day when man and machine become one, we should ask if we're truly passing along the best aspects of ourselves -- the limitless nature of our soul The resiliency of our spirit And the endless possibilities of a mind that has been set free.