Return to Transcripts main page

INSIDE MAN

Morganbot

Aired January 22, 2015 - 21:00   ET

THIS IS A RUSH TRANSCRIPT. THIS COPY MAY NOT BE IN ITS FINAL FORM AND MAY BE UPDATED.


MORGAN SPURLOCK, CNN ANCHOR: Robots. They've been promised to us for generations, that autonomous, thinking machines will revolutionize our work, our homes, and everything in between. Today, technology is advancing so rapidly that we might finally be on the cusp of the new age of robotics and artificial intelligence that we have dreamt about for so long.

The question is are we ready for them and are they ready for us?

Give me a call. I just want to catch up on two shows.

It's time to start working on a new episode of "Inside Man".

Good morning. Good morning guys.

This week's episode is all about the world of robotics and artificial intelligence. I'm excited to hit the road and start filming.

What's happening today?

UNIDENTIFIED FEMALE: 10:00 A.M. call Lizzie and Benny (ph)...

SPURLOCK: Only problem, my life is kind of getting in the way. We got projects in development.

UNIDENTIFIED MALE: Moving things along with the fight (ph) church deal.

SPURLOCK: Episodes of Inside Man, we're editing.

How far into the episode are you right now?

And we're finally updating our website.

UNIDENTIFIED MALE: It's time to have a web presence.

SPURLOCK: And I have meetings, upon the meetings, upon the meetings.

UNIDENTIFIED FEMALE: Do you want to see the beginning of the disregarded (ph) shows?

SPURLOCK: I'm always taking on a law but this time I may have gone too far.

UNIDENTIFIED FEMALE: You're going to be in D.C. for the day. UNIDENTIFED FEMALE: Alaska's...

UNIDENTIFIED MALE: Nebraska.

SPURLOCK: How am I going to get everything done and shoot this episode?

OK.

The answer might just be putting those two things together.

That's right, I'm going to harness the power of artificial intelligence to help me become more productive and get it all done.

First step, look for that perfect robot helper.

I mean, let's be honest, a robot of my dreams is probably the robot of your dreams. This humanoid robot, that can walk around, talk like us, do things for us, you know, everything that we would imagine it could do in the real world. But, in the real world these types of robots are few and far between. But I'm about to go and meet Dennis Hong and if anybody can help me find my robot, it's this guy.

DENNIS HONG, PROFESSOR AND THE FOUNDING DIRECTOR OF ROMELA: Hey, Morgan.

SPURLOCK: Hey.

HONG: How are you?

SPURLOCK: How are you?

HONG: Thanks for coming.

SPURLOCK: Thanks for having me.

HONG: Yup. This is the robotics lab, RoMeLa and this is where the magic happens.

SPURLOCK: With 20 robotic creations under his belt, Dennis Hong is one of the world's leading roboticists.

Is there a choice you make when making robots to make them more robotic looking versus more human looking?

HONG: If it requires the human form and then we build a robot in a human shape. Walking with two legs is very, very difficult. If it doesn't need two legs, there is no reason to put two legs on it.

SPURLOCK: Why do we need two legs?

HONG: It can use to move around. In your home, there are stairs. When you open the door to door handle at a certain height, why, because, it's designed for us humans to use.

SPURLOCK: Right.

HONG: So on that the robot is a human shape and size. It won't be to navigate the environment.

SPURLOCK: Who's that guy hanging out over there?

HONG: Yeah. This is a robot called Darwin. It stands for Dynamic Anthropomorphic Robot with Intelligense.

DARWIN, ROBOT: Hello everyone. My name is Darwin. I'm open architecture, a humanoid platform for research education and outreach activities.

SPURLOCK: But Darwin isn't just a cute research robot, he's also a world class soccer champion. That's right. Darwin and Dennis Hong are gold medal winners of RoboCup, the Robotic Soccer Tournament and it turns out that soccer field was the perfect place to whom the skills of robot needs to operate autonomously.

HONG: You can see it actually sees things, right.

DARWIN: Right.

HONG: The actual soccer match is much more as complex as you can imagine. He needs to know where he is in the soccer field.

SPURLOCK: And where the other players are...

HONG: Other players are and then he needs to think again and predict.

SPURLOCK: Right.

HONG: He has to understand the game of soccer itself.

SPURLOCK: Goal.

They're the same skills that robots will need out in the real world in order to recognize and understand their surroundings and interpret interactions accurately, create goals and act accordingly. Things that are second nature for you and me but monumental challenges for robots.

HONG: These robots work in their really very controlled environment. In the lab other robots do perfect soccer. You take it out of the lab and put it at a situation, he gets confused. So try to make a robot that can really deal with uninstructed environment. Outdoor is very, very challenging.

If you want to use robots for real life applications, disaster relief -- it needs to walk over rubble...

SPURLOCK: OK.

HONG: ... it needs to climb unstructured stairs and those kinds of things then we still have a long way to go.

SPURLOCK: What's next?

HONG: We also have hexapod robot called Harry.

SPURLOCK: So the hexapod robot are cool but this one it starts to get creepy because it's like spider robots...

HONG: Yeah.

SPURLOCK: ... that's coming to you like to take over. What will be the applications for a robot like that?

HONG: So this could be used for a very unstructured rock terrain where wheels cannot go and sometimes two legs doesn't cut it.

SPURLOCK: Yeah.

HONG: So we have six legs. It can climb up different surfaces and also it's always stable.

SPURLOCK: What will be the first robots that are going to happen?

HONG: Medical robots, military applications, disaster relief. Where human lives are in stake, that's where you're going to see robots being used.

SPURLOCK: That all sounds great but I came here to find a robot that can help me in my life today and the jury is still loud about how much this little robots can do out in the real world.

Do you think we can trying put the robots through some personal trials with me to see what they can actually do?

HONG: Yeah, we can do some fun experiments.

SPURLOCK: The first thing is first -- lunch.

Good boy, Harry. Thank you very much. Who's a good robot?

So far and so good, this could be the beginning of a beautiful friendship.

But what about some more complicated tasks like helping me with work?

Darwin, remind me to call the office.

DARWIN: Yes.

SPURLOCK: It's like written and cling on. And it could jump right on the wall next time, how about like a piece of paper.

Or, when I find myself in a comprising situation.

Oh come on. Harry, bring me some toilet paper. Over here. Come on. Really? What I'm going to do with this? Go get a fresh roll. Come on, hurry up. Robots. All right, so maybe I might have to wait a few years for a personal robot helper. But something does tell me when these robots finally do come around, it's going to be pretty amazing. But just because I can't use a humanoid robot right now, it doesn't mean robotics can help me with some of the things in my life. But the thing I just have to figure out is how.

(COMMERCIAL BREAK)

SPURLOCK: I'm on the road shooting our new episode of Inside Man. All about robots and artificial intelligence and to help me deal with my busy Cadillac (ph) back home, I've decided to use this opportunity to get some much needed help.

Remind me to call the office.

I've learned that we are a few years away from a having a real world robot helper.

What I'm going to do with this?

But I'm not giving up hope that there's a robot out there that will help me to be more productive.

HONG: He might kick you.

SPURLOCK: And where better to find my dream robot than Japan.

The Japanese love their robots.

There's no place where you feel that impact more than here in Tokyo. When you're in the middle of the city, you actually do feel like one day robots will take over.

Here, everything is seems to be influenced by robotics. Food service Look at what I got. Public restrooms, even the toilets are robots. And then there is this guy.

Ever since I was a little kid, these were the robots that imagine, the giant robots, superheroes are protecting us and taking care of the world and here in Japan they share the same passion I did.

Japan's love of robots can be traced to something called Sansuikan the belief in Japanese culture that everything has a presence. Human and robot are alike. And nobody is playing with the boundary between human and robot more than Dr. Hiroshi Ishiguro, creator of the most realistic Androids ever made.

And no, he doesn't have a twin at least not a biological one. That's Jimanoi (ph), the android replica he made it himself.

Ishiguro's goal to create a sort of robot double for himself that could function in his place and essentially allow him to be in two places at once. Something at this moment sounds very appealing to me in my busy schedule.

Ishiguro's latest android creations are on display here at Miraikan -- the National Museum of Emerging Science and Innovation.

They're part of an exhibit called "Androids, What is Human," which invites the spectator to explore the differences between us humans and our android counterparts.

Ishiguro believes that a robot can have a presence that's indistinguishable from that of a real human.

This is Kodomoroid the child newscaster who recites news reports gathered from around the world in a variety of voices and languages.

It's amazing to see you like up close. It is (inaudible).

I think she have a little episode right there.

Otonaroid is a teleoperated android resembling in adult woman.

Machi (ph) my translator in a booth and I'm going to be talking to the android.

Hi Machi.

UNIDENTIFED FEMALE: Hello.

SPURLOCK: Tell me about the best part of today.

UNIDENTIFED FEMALE: It's being android. This is my first day experience. So...

SPURLOCK: As Machi operated from the booth, Otonaroid translates her voice and mimics her movements.

(Inaudible) it's so weird.

UNIDENTIFED FEMALE: I don't know. I'm too shy. I'm Japanese.

SPURLOCK: But you can turn your eyes when you look at me.

These robots plunge deep into what's known as the uncanny valley -- hat feeling of revulsion that people feel when interacting with something that looks human but not quite.

That's creepy.

UNIDENTIFED FEMALE: Do you feel comfortable talking to me.

SPURLOCK: I'm feeling weird because what happen is like there will be things that make the android look incredibly human like when you move the eyes, yeah, but them those things they kind of give away the facade like the movement of the mouth or the movement of the arms. The minute all of those things are fixed it will be unbelievable.

The telenoid -- what are the very minimum fundamental requirements for human appearance? The telenoid was created to understand this question. This 3,000 percent more creepy.

Meet telenoid, another teleoperated android.

Hello. Oh my gosh, this is so like a freaky baby.

TELENOID, ROBOT: Hi Morgan.

SPURLOCK: Hi. How are you?

TELENOID: I'm OK. (inaudible)

SPURLOCK: I'm like in the movie Cacoon right now. This is like preparing us to like an alien invasion. Say hello to everybody at home.

TELENOID: Hello everyone. How are you all?

SPURLOCK: Waive.

What would be interesting is when you don't need a person to operate it like when basically it will be able to almost like interact with you and talk to you without needing to be driven by someone because I think that's -- I don't know -- more useful.

What do you think?

TELENOID: I think it's good idea.

SPURLOCK: Weird. So weird. You are going to give me nightmares tonight. Yes. Yes.

TELENOID: I'll be hugging you.

SPURLOCK: Nightmare, nightmare.

TELENOID: No nightmare.

SPURLOCK: Yeah, even my dreams like you're going to come hoping in, oh hi Morgan. Oh, I love you and you're going come up and then you're going to try to strangle me with your little arms.

Amazing? Yes. Creepy? Absolutely.

So maybe we're not quite at the point where I can have a Morgan Spurlock android to take my place in business meeting and speaking engagements but when it comes to virtually being in two places at once, all hope is not lost.

I'm about to meet a guy who just might have the answer and the helper that I've been looking for.

SCOTT HASSAN, PRESIDENT, SUITABLE TECHNOLOGIES: Welcome to Suitable Technologies.

SPURLOCK: Thank you Morgan.

Scott Hassan is the President and CEO of Suitable Technologies located in Palo Alto, California. We're there on the cutting edge of a new way of robotics.

I need a little help. I'm constantly needing to be in more than one place at one time and I was hoping you can help me with that.

HASSAN: Well, here at Suitable Technologies we make the same called a Beam.

SPURLOCK: OK.

HASSAN: It's a -- we call a Smart Presence System. It allows you to be anywhere, anytime.

SPURLOCK: What is it do?

HASSAN: So it's like Skype except that it's Skype plus ability to move.

SPURLOCK: What are the biggest uses right now?

HASSAN: Mostly corporate companies...

SPURLOCK: Yeah.

HASSAN: ... that have multiple offices so they basically they need now beam from one office to another officer.

SPURLOCK: So how do you think a beam can help me?

HASSAN: So for instance, you could do -- have done this whole segment...

SPURLOCK: Yeah.

HASSAN: ... going to be where you beamed in here where, you know, there's no travel time for you. You just click and you probably could do four or five days a day.

SPURLOCK: I love that.

HASSAN: But let me show on a little car.

SPURLOCK: I have to say I'm impressed. The beam's design is smart, simply and functional. And for a cool $16,000 it can be yours.

This is where you would beam in somewhere else in the world?

HASSAN: This is where I can test beam from here.

When was the last time when you're in Paris?

SPURLOCK: It's been awhile.

HASSAN: Beam into Paris right here.

SPURLOCK: OK. Oh wow, check that out. That's pretty amazing.

Hi, there. Hello.

(FOREIGN LANGUAGE)

UNIDENTIFIED MALE: Hi.

SPURLOCK: Hi. Bonjour. Where you guys from?

(CROSSTALK)

SPURLOCK: Are you guys on vacation?

UNIDENTIFIED MALE: Yeah, we are.

SPURLOCK: I'm on a virtual vacation.

UNIDENTIFIED MALE: It is nice to meet you.

SPURLOCK: It's wonderful to meet you guys. Bye guys.

UNIDENTIFIED MALE: Same here.

: Cool. So where should we head to now?

HASSAN: Have you been to Vegas?

SPURLOCK: I love Vegas.

Hello, there.

UNIDENTIFIED FEMALE: Hi.

UNIDENTIFED FEMALE: Hi.

SPURLOCK: How about we line up some drinks here.

UNIDENTIFIED FEMALE: So we have just 88 ounce daiquiri.

SPURLOCK: Yeah, I think I need to do that.

Only in Vegas, Ma'am -- only in Vegas.

Of course actually drinking that daiquiri would be nice but then again it's never been easier to skip out on the bill.

HASSAN: Finally, we have one last place to beam into.

SPURLOCK: This is a mystery.

HASSAN: Yeah.

SPURLOCK: And...

HASSAN: Do you recognize that place?

SPURLOCK: I know exactly where we are. This where I grew up That's awesome. This is my room as a kid. That's amazing. It feels like I'm actually

home. Hey, who's that in the kitchen?

PHYLLIS SPURLOCK, MORGAN SPURLOCK'S MOM: Hi.

SPURLOCK: Hey, mom.

P. SPURLOCK: Hi.

M. SPURLOCK: Hey, Addison, how are you?

UNIDENTIFED MALE: Good.

P.SPURLOCK: Nothing too exciting going on.

M. SPURLOCK: Mom, you're talking to a robot. That's exciting.

P. SPURLOCK: That's questionable.

M. SPURLOCK: For somebody like me who doesn't get to get home as often as I'd like to, it's like the next best thing.

P. SPURLOCK: I wish you were here.

M. SPURLOCK: You can still come give me a hug.

P. SPURLOCK: OK.

M. SPURLOCK: Then give me a big hug.

P. SPURLOCK: And now I'll give you a hug.

M. SPURLOCK: Give me a big kiss.

P. SPURLOCK: OK.

M. SPURLOCK: Thanks, mom. I love you.

P. SPURLOCK: I love you, too.

M. SPURLOCK: All right. I'm sold. That's it. Done. They had me and my mom hugging the computer. I'm getting one of those on my office immediately to be able to travel around the world and have a communication with my office back in New York City is going to be epic. I can't wait. So done.

(COMMERCIAL BREAK)

SPURLOCK: Hey, guys. Hey, there. How are you, brother?

UNIDENTIFIED MALE: Mr. Spurlock, how's the Seattle?

SPURLOCK: It's not raining.

My Beam has finally arrived in my office and although I'm across the country right now, it's as if I've never even left New York. How is the edit going?

UNIDENTIFIED MALE: It's going pretty fantastically.

SPURLOCK: I can even make good on that promise to meet with my team about our new company website.

I thought we should get together while I'm out of town and have a little website conversation.

UNIDENTIFIED MALE: I think we're down to the last few things. We need to settle on our layout.

UNIDENTIFED MALE: You need to write the bios. We need to populate our frequently asked questions.

SPURLOCK: Yeah, I can do that. Like I give you the bios, and I'll do this frequently asked questions.

My plan to use robotics just might work out after all.

Stacy (ph)?

Being in two places at once, check.

Today we're going to Microsoft. We're going to talk to one of the geniuses who behind A.I. there.

UNIDENTIFED MALE: (Inaudible)

SPURLOCK: Do it tomorrow.

UNIDENTIFIED MALE: Are you going to come to the office or is this going to be a permanent fixture in our lives?

SPURLOCK: I kind of like this. See? Now, I'm going to storm away. So take that.

UNIDENTIFED MALE: I...

SPURLOCK: And you guys get to work.

Robots are slowly but surely inching their way into our lives. But at the heart of every robot is its brain, its artificial intelligence or A.I. That allows it to make autonomous decisions about the world around it and act accordingly. A.I. is still from the early stages, but futurists, computer scientists, you know, sci-fi enthusiasts like me, all believe that one day, artificial intelligence will either catch up to or surpass human intelligence. And at that point, you know, all bets are off. Anything's possible.

But I want to know how A.I. is going to affect our lives in the near term. To find out, I'm in Washington State to visit one of the oldest players in the computing game.

And from the moment I set foot in the Microsoft Research Headquarters, things seem a little different, an elevator that seems to know when I'm coming...

ELEVATOR ROBOT: Going down.

SPURLOCK: A robot ready to help and guide me through the building.

DIRECTIONS ROBOT: Eric Horvitz's office is 30126. Turn left and go straight to the end of the corridor. By the way, I'll let Monica, his virtual admin., know that you're on your way.

SPURLOCK: And instead of a human assistant in front of A.I. researcher Eric Horvitz's office, there's a virtual one.

MONICA, ERIC HORVITZ'S VIRTUAL ADMIN ASSISTANT: Hi. I was expecting you. The robot told me you were coming. Are you here looking for Eric?

SPURLOCK: Yes, I am.

MONICA: Are you Morgan Spurlock?

SPURLOCK: I am.

MONICA: Great. I see that your meeting is on his calendar. If you don't mind waiting around for a few more minutes, I'm sure he'll invite you in shortly. I guess, I'll see you later then. Bye-bye.

SPURLOCK: Thank you. It's amazing. See, now I want it to turn and talk to me, though. Would you like a glass of water?

ERIC HORITZ, MICROSOFT A.I. RESEARCHER: Hello, there.

SPURLOCK: Hey, how are you?

HORITZ: Hi, nice to meet you, Morgan.

SPURLOCK: Good to see you. I was talking to your virtual assistant.

Eric has been developing his virtual admin., also know as Monica, for over five years. And although my interaction with her seemed simple enough, there's a lot more happening in her artificial brain than meets the eye.

HORITZ: She actually was doing her best job to protect me just now. She'll always checks to see what's the cause of me being interrupted at any moment. She watches me, she's my best cop. She sees me interacting with content, typing, and decides that the cause of interruption now is just probably too high, so hang on one second.

SPURLOCK: In other words, she's doing what a human assistant would do -- leveraging all the information at her fingertips to make game time decision.

Monica, the directions robot, and the elevator, are all part of the Situated Interaction Project, a research project that enables humans and machines to have complex social interactions. HORITZ: It's all about the idea of taking intelligence systems and

giving them knowledge of the context to the physical surroundings and the social context of who's there, and who's talking to who, and relationships in the world, and they combine multiple strings of information, some vision, they're looking for people, understanding the locations.

I have to go to the bathroom.

DIRECTIONS ROBOT: The bathroom is just down this hallway on your right.

HORITZ: OK. They can recognize speech. They actually know where the speech is coming from. They can do a multiparty and ask questions of several people in a group. And also reason, determining, "Does this person want to be engaged to me right now or not?"

SPURLOCK: So basically, the systems will analyze my intention?

HORITZ: Yeah. Intentions are often hidden inside people. But we -- our behaviors and our speech, just to communicate with others typically people about what it is we want.

SPURLOCK: In order to teach machines how to interpret human intention, Eric and his team used a bedrock of A.I. called, "Machine Learning." Overtime, the elevator learns to distinguish the body language of someone who needs a ride versus somebody simply walking by.

HORITZ: The system gets smarter and smarter and smarter continually and it becomes very right about predicting and ordering the elevator in advance of people even coming to the elevator.

SPURLOCK: And back in Eric's office, Monica is ready with an update.

MONICA: No one has stopped by to see you since we last talked. Catch you later.

HORITZ: OK.

MONICA: Bye-bye.

HORITZ: Thanks.

SPURLOCK: That's incredible. One of the things that I love about my assistant is that she can actually have, you know, real conversations with people. How long until something like that is kind of going to be available in this type of a form?

HORITZ: We have projects on chitchats and conversation that are underway.

SPURLOCK: Yeah.

HORITZ: So I'd say within a couple of years.

SPURLOCK: We actually got that much time.

What I'm realizing is that the ability for machines to make intelligent decisions and complete cognitive tasks on their own is much more valuable than a robot that can do a physical tasks for us, and the best part is, is that's happening right now. We have access to that technology today.

So the question is now, how can I harness that technology, that artificial intelligence to help with me with my ever growing to do list?

(COMMERCIAL BREAK)

SPURLOCK: My visit to Microsoft has got me thinking, A.I. programs that can assist us with cognitive tasks are just as useful as robots that can help with physical tasks.

HORITZ: We have projects on chitchats and conversation that are underway.

SPURLOCK: And now that I know how smart artificial intelligence can be, I'm hoping it can lend a hand with my growing to do list.

UNIDENTIFIED MALE: We need to populate our frequently asked questions.

SPURLOCK: So I'm back in California to meet with some folks who may be on to something that can help.

Hey, I'm Morgan.

BRUCE WILCOX, ARTIFICIAL INTELLIGENCE PROGRAMMER: Hi. How are you doing?

SUE WILCOX, ARTIFICIAL INTELLIGENCE PROGRAMMER: Hello. Nice to meet you.

SPURLOCK: The Bruce and Sue Willcox are computer programmers specializing in chatbot -- A.I. programs designed to carry out intelligent conversations with humans.

I am trying to find some type of A.I. that will alleviate some of the work I have to do in my life. I told my team back in New York that I would write a frequently asked questions section to my website and I thought you guys might be able to kind of help me create something like that.

B. WILCOX: Nowadays, chatbot are -- out of the things you see in corporate websites were Siri -- I mean, a lot of companies have a corporate (inaudible) on their website.

JENN, ALASKA AIRLINES VIRTUAL ASSISTANT: I'm Jenn, Alaska Airlines automated travel assistant.

B. WILCOX: Ask good questions about the company or about its products, it would just answer you. Chatbot on a broad scale are these interactive conversational things where when you start talking, the goal is you shouldn't be able to tell it from a human being.

SPURLOCK: That would be the ideal. Maybe we make them think that it's me that they're having this conversation with.

B. WILCOX: We normally create a character from scratch...

SPURLOCK: OK.

B. WILCOX: ... where this is unusual for us to because you exist.

SPURLOCK: Right.

B. WILCOX: And we're going to try and replicate you. When do you want by?

SPURLOCK: Yesterday.

B. WILCOX: OK. That's going to be a little bit harder to do.

SPURLOCK: Yeah. How many years from having a completely truly interactive A.I. like in the movie Her?

B. WILCOX: We're so far from that A.I. capability. Right now it's 2029.

SPURLOCK: OK.

B. WILCOX: Like a lot of government contracting projects, right? Every 20 years they say, "OK, we didn't do. It will be in next 20. We promise."

SPURLOCK: How do you make a chatbot?

B. WILCOX: We're going to try and learn as much as we can about you and put that into the chat button. We know what you say in response to certain questions. So now what I'm trying to do is replicate in the chatbot, recognizing those questions.

SPURLOCK: So how long do you think it will be until I have some of a Morganbot?

B. WILCOX: We're going to be aiming to have it done in the next couple of weeks.

SPURLOCK: Awesome.

Mission accomplished. Now, I don't have to use this at all. Something I'm already really good at. They're going to get work on my chatbot in just a matter of time. You're going to go the website and talk to my Morganbot. Done and done.

That's right. This is going to save me a big chunk of time and allow me to multitask instead of being headed to a computer.

But what about the time I'm still wasting behind the wheel of my car up to an hour a day? How can I make that more productive? How are you doing? I'm Morgan.

CHRIS GERDES, CENTER FOR AUTOMOTIVE RESEARCH AT STANFORD DIRECTOR: Chris. You ready?

SPURLOCK: Chris Gerdes may have the answer. Since 2009, Chris Gerdes and his team at Stanford University have been working on their self- driving race care also known as Shelley.

How fast has Shelley driven all by herself?

GERDES: Probably somewhere around 135, 140 miles an hour.

SPURLOCK: That's pretty fast.

As it turns out, a car pushed to the limit on the race track is a window into the type of stress a regular car encounters in places, and it's a great way to figure out what it takes to regain control.

GERDES: So our ideal is we can make a self-driving car that would have the same skills as the best human race car drivers. We could really go a long way towards improving safety.

SPURLOCK: Cool. Let's check out the car.

GERDES: All right. Our test is in the palm of your hand.

SPURLOCK: OK. This is it. Ready? There's nothing quite as scary than watching the wheel turn by itself.

GERDES: People tend to be terrified initially.

SPURLOCK: And so it's knowing how far out it needs to go to make this turn?

GERDES: That's right. It's actually taking the racing line like a driver would. Here's the fun one up on the top of turn 5. She has to break, turn, and then accelerate.

SPURLOCK: Wow. All on a hill. That's incredible. And you can feel it's starting to get that little bit of grip as it goes down that turn.

GERDES: That's right. You could feel the rear end sliding and her correcting a little.

SPURLOCK: Shelley recalculates the corrections at an astounding 200 times a second with the help of a laser scanner, which finds the edge of the road and looks out for obstacles along the way and two GPS antennas that communicate speed and the exact position on the track.

I want to go faster.

GERDES: Safety first. As I say in racing, if you want to finish first, first you got to finish.

SPURLOCK: Three miles of flawless driving, and Shelley returns us safe and sound to pit road.

That was phenomenal. I made it. I lived. I didn't get smash into the wall by a robotic car. If Shelley has any indication of where the future is heading, I mean it's only a matter of time before driverless cars or really seriously A.I. assisted cars are just on the road full time. And that type of robotics gets me very, very excited.

Being able to purchase a self-driving car maybe a few years away, but when they hit the market, I'll be the first in line.

(COMMERCIAL BREAK)

SPURLOCK: My plan to use robots in artificial intelligence to help get my work done is off to a good start.

I have my telepresence robot. Yey, website. And the creation of my chatbot is underway but I still have to write that bio from my website.

The question is, "Can I actually find a writerbot that will do this for me? Can artificial intelligence do the work so I don't have to?"

KRIS HAMMOND, NARRATIVE SCIENCE CHIEF SCIENTIST: Hey, how are you?

SPURLOCK: I'm Morgan.

HAMMOND: I'm Chris.

SPURLOCK: Kris Hammond is Chief Scientist here at Narrative Science -- a Chicago based company with offices here in New York City.

Tell me about your company. I'm fascinated by the whole thing.

HAMMOND: So we've taken the entire world of data -- big data, numbers, symbols, grafts, charts, and we've turned into English.

Well, our system can pull that story out and turned it into something that is readable by any audience you can imagine.

SPURLOCK: One of the things I need to do is write my bio for our company website and rather than me do it, I'm looking for something else or someone else that could help me do it and so I don't have to.

HAMMOND: That what Quale (ph) is designed to do. It's designed to take your life and make it easier.

I think running a bio for you makes tremendous sense.

SPURLOCK: Perfect. You're hired.

HAMMOND: Awesome. Thank you.

SPURLOCK: Quale (ph), their A.I. platform is going to write a bio for m based on the information and access is from the Internet Movie Database and online directory for film and television credits. HAMMOND: This is actually what our content architects see. And what we do is we can configure system where we teach about what is important and what will be interested. And we tell which kind of story it was.

Once we've done all of that, we have preview. And that's it.

SPURLOCK: So, it is taken all that data and like that.

HAMMOND: Literally, written that story in real time.

SPURLOCK: I want the first line. He's a writer, director, and producer but might be most famous from being himself.

HAMMOND: And being -- I have tell you how exactly why that is? From IMDb's point of view, more than anything else, your role is at yourself.

SPURLOCK: Yeah, we go to the credits and listed as himself.

HAMMOND: That's exactly right.

SPURLOCK: So -- same as for being himself.

HAMMOND: For me, that was like -- that's actually....

SPURLOCK: I can get a human whoever write it like that which I think is interesting and compelling.

Spurlock is an accomplish filmmaker with credits in 81-TP projects including T.V. appearances in 39 films. A little known fact about Spurlock, he is rejected by the Film School of USC five times.

HAMMOND: God bless. And...

SPURLOCK: That's right -- yeah.

There's a few things in here that I would tweak and change but on a whole it's pretty amazing.

This was completely written by a machine.

Well, the machine never win a Pulitzer.

HAMMOND: Absolutely.

SPURLOCK: Yeah.

HAMMOND: Oh no, I'm happy to go on record with that.

You know, we hear about how there will be a time where computers like should be able to think for themselves, make their own conscious decisions. This is the first step toward something like that.

And this is the beginnings of that because if you think of what artificial intelligent is supposed to be, it's supposed to be -- it's a model of human thought and skills. And what is the one thing that everybody says is the thing that separates us from animals. It's language.

And this can generate language in a way that no one has ever had a machine do it before. And that's super exciting.

SPURLOCK: But if Kris thinks an A.I. program and win a Pulitzer one day, would that also make documentaries and win an academy award. And if so, what is that mean for me?

I'd be lying if I said I wasn't amazed by the technology I've seen. And it's truly mind-boggling to think of all the ways these advancements will impact our lives in the future. But its robots and computer programs continue to take over jobs that will once the domain in humans, what will it mean for society?

It's not the first time we had to grapple with these questions. After all, the industrial revolution forever change the course of human civilization.

ERIK BRYNJOLFSSON, MIT PROFESSOR: Good to see you.

SPURLOCK: Morgan.

BRYNJOLFSSON: Erik.

SPURLOCK: An MIT professor, Erik Brynjolfsson argues that these changes will be justice radical. He calls what we're embarking on the "Second Machine Age".

Compare kind of where we are today...

BRYNJOLFSSON: Yeah.

SPURLOCK: ... and the industrial revolution.

BRYNJOLFSSON: In the First Machine Age, we automated and augmented our muscles, physical power. In the Second Machine Age, we're doing a lot of the same things for our brains, cognition.

And so, when the First Machine Age kicked in humans became more and more valuable because of their brains, but now if the machines can do both the physical power and the mental parts of the tasks, you know, what's our role?

SPURLOCK: Yeah.

Are there going to be jobs and careers that are kind of machine age proof?

BRYNJOLFSSON: You know, I wish I could say, yes, for sure. Right now, there's been a big drop in demand for routine information processing jobs.

Something like 60 percent of Americans are primarily doing information processing. You know, clerks, bookkeepers, accountants, a lot of those tasks can be done better by machines. And the big challenge is, can we invent the next new jobs to replace the old ones?

SPURLOCK: Part of the problem people can't adopt quick enough.

BRYNJOLFSSON: Yeah. I'm not ready to say we can't. The problem is the technologies advancing is so quickly now...

SPURLOCK: Yeah.

BRYNJOLFSSON: ... that we just not keeping up.

SPURLOCK: What is the worst case scenario of what can happened in the next machine age?

BRYNJOLFSSON: Well, you get about as bad as you can imagine.

Things we should be probably worrying about right now?

SPURLOCK: Yeah.

BRYNJOLFSSON: I think this growing in equality could accelerate a lot. Where -- yes, the pie get bigger where it becomes much more concentrated and 57 percent, 70 percent, 90 percent of Americans end up being worst off.

SPURLOCK: It's easy to see how through our excitement for technology, we can create more and become disparity. But when you also start realizes is as there's more and more advancements, everyone could easily become irrelevant.

So what do we do to prepare for that? How do we make sure that doesn't happen? And what can we do moving forward to make sure they were prepared enough to be blind-sided by that is a possibility.

(COMMERCIAL BREAK)

SPURLOCK: I'm using the power of robotics and artificial intelligence to help me make me more efficient and productive in my busy life.

UNIDENTIFIED MALE: Open the five (ph) big doors now.

UNIDENTIFIED MALE: I'm afraid I can't do that.

SPURLOCK: When we've all seen the movies where A.I. turns on humans, and according to recent statements by top technologists and scientists, these doomsday scenarios are not just the stuff of science fiction.

UNIDENTIFIED MALE: I'd like to just keep an eye on what's going on with artificial intelligence. I think there is potentially a dangerous outcome there.

UNIDENTIFIED MALE: Artificial intelligence could be a real danger in the not too distant future.

SPURLOCK: Or even in a not too distant past.

In 2009, Swiss researchers made the startling discovery that over time, robots stay at program to share our food source of one another, had in fact started to selfishly hoard it for themselves even though no one had told them to do so.

The robots had an essence gone broke. So our folks Eric Horvitz who were actually developing artificial intelligence's concerned.

What is the future of A.I.? We read the Sci-Fi books. I see the movies where A.I. goes crazy and the next thing, you know, machines are coming to attack human beings and it's the end of the world.

HORVITZ: You know, any powerful technology can be used in great ways and in ways that for people and society and we've seen that forever.

SPURLOCK: Yeah.

HORVITZ: The same can be through of A.I. And this is something that scientist like myself and others deliberate about.

SPURLOCK: Because right now, there's a problem we can turn it off. Will they ever not let us turn them off?

HORVITZ: And these are the kinds of scenarios that we have been thinking about that someday there could be a superintelligence that was so bright we could understand it and intelligence explosion.

So, how can we design systems for example that human beings could always turn off no matter what. The idea basically is to provide proactive guidance and a potential concerns like this.

SPURLOCK: Yeah.

HORVITZ: I think they're real actually. I don't think -- I don't want to...

SPURLOCK: You can't trivialize them.

HORVITZ: I don't want to trivialize them.

I happened to be extremely optimistic about things. Where does any systems fight their very nature that have our deep interest in mind.

We encode, we can view them -- appreciation of our preferences. So, in short, I believe we are building systems to all come to love us and help us and compliment us.

SPURLOCK: That's the hope.

HORVITZ: That's the hope in that and my belief.

SPURLOCK: Yeah.

I'm back in New York just in time to try out my new chatbot. The online persona that will answer the frequently asked questions on our new website.

So the programmers that I met in California, Bruce and Sue Wilcox had been working diligently on the Morgan chatbot.

They have been sourcing interviews I've done, questions I've answered, any information that exist out there about me into this program that is living online and it's going to give you experience that you're talking to me.

So now for the first time ever, I'm going to talk to myself on line.

How are you?

I'm pretty good. How are you?

Where are you from?

I was born in the country in Parkersburg, West Virginia.

What's your favorite movie?

Facing the Crowd because it's such a great story about the power of demigods in T.V.

Good answer. Where are you now?

Right where I am. It's like I'm yoda. Right where I am, I am.

What's your favorite food?

Do you like cleaning up after people?

Turkey with gravy because it reminds of mom's cooking.

It's like I'm talking to myself which like, you know, some sort of like verbal torrent (ph). And kind of like -- it's just like, Oh, (inaudible).

Do you have any brothers or sisters?

I have two older brothers.

What do they do?

They remember non-verbs.

Hey, Craig Spurlock?

CRAIG SPURLOCK, MORGAN SPURLOCK'S BROTHER: They remember non-verbs. What does that mean?

M. SPURLOCK: I don't know. You tell me. What do you do?

You remember non-verb.

C. SPURLOCK: But usually I'm remembering verbs. They got that part wrong.

M. SPURLOCK. I'm going to tell them.

It's a pretty amazing thing to get to have a conversation with some version of yourself and it is a valued-effort. You know, where this is, is impressive because you see a little bits and moments of what is essentially your personality or things you've said or something that really does connect you to the real world and a way that I find to be really unique.

But at the same time, it's still a robot. And as cooled as this like be our action is. This interaction talking of people is something that I have found much more valuable.

So why give that up? I think there's room for both. I mean, robotics and A.I. are already around us every minuet of everyday...

UNIDENTIFIED FEMALE: Recalculating.

SPURLOCK: ... in ways where barely even conscious of. And so far, we've manage to use them to our advantage.

UNIDENTIFIED FEMALE: I'll remind you to pick up flowers when you leave work.

SPURLOCK: I was, in fact, able to harness the power of both to help me be more productive, to accomplish more in a day and to be into places at one at least virtually.

UNIDENTIFIED MALE: How's Seattle?

SPURLOCK: It's not raining.

And maybe one day, we will be facing about machines that can help perform us, outsmart us.

UNIDENTIFIED FEMALE: Are you Morgan Spurlock?

SPURLOCK: And maybe even out human as humans.

That's creepy.

But until then, for better or worst, this planet is ours. So let's enjoy it while we're still in-charge.

UNIDENTIFIED MALE: Let's party in the office. Eric, Eric, Eric I'm right behind you dancing in the office. And I see you and I'm dancing in the office. And now you don't see that I see you in the office. I'm going around and around and around and around.

SPURLOCK: Here's come baby Baltimore (ph) hopping into town. Here's come little baby Baltimore. Look at him fall down.