0:00:00.564,0:00:03.086
My job is to design, build and study

0:00:03.086,0:00:05.840
robots that communicate with people.

0:00:05.840,0:00:06.639
But this story doesn't start with robotics at all,

0:00:06.639,0:00:08.531
it starts with animation.

0:00:08.531,0:00:11.117
When I first saw Pixar's Luxo Jr.,

0:00:11.117,0:00:13.103
I was amazed by how much emotion

0:00:13.103,0:00:14.861
they could put into something

0:00:14.861,0:00:17.452
as trivial as a desk lamp.

0:00:17.452,0:00:19.285
I mean look at them -- at the end of this movie,

0:00:19.285,0:00:21.789
you actually feel something for two pieces of furniture.

0:00:21.789,0:00:23.617
(Laughter)

0:00:23.617,0:00:25.737
And I said, I have to learn how to do this.

0:00:25.737,0:00:29.271
So I made a really bad career decision. (Laughter)

0:00:29.271,0:00:31.675
And that's what my mom was like when I did it.

0:00:31.675,0:00:33.686
(Laughter)

0:00:33.686,0:00:35.866
I left a very quasi-tech job in Israel

0:00:35.866,0:00:38.471
with an iSoftware company and I moved to New York

0:00:38.471,0:00:39.482
to study animation.

0:00:39.482,0:00:40.639
And there I lived

0:00:40.639,0:00:43.635
in a collapsing apartment building[br]in Harlem with roommates.

0:00:43.635,0:00:45.003
I'm not using this phrase metaphorically,

0:00:45.003,0:00:46.963
the ceiling actually collapsed one day

0:00:46.963,0:00:48.273
in our living room. (Laughter)

0:00:48.273,0:00:50.939
Whenever they did those news stories [br]about building violations in New York,[br]

0:00:50.939,0:00:53.185
they would put the report in front of our building. [br](Laughter)

0:00:53.185,0:00:56.783
As kind of like a backdrop [br]to show how bad things are.

0:00:56.783,0:00:58.729
Anyway, during the day I went to school and at night

0:00:58.729,0:01:02.251
I would sit and draw frame-by-frame [br]of pencil animation.

0:01:02.251,0:01:04.525
And I learned two surprising lessons --

0:01:04.525,0:01:07.078
one of them was that

0:01:07.078,0:01:09.038
when you want to arouse emotions,

0:01:09.038,0:01:10.764
it doesn't matter so much how something looks, [br]

0:01:10.764,0:01:13.110
it's all in the motion -- it's in the timing

0:01:13.110,0:01:14.717
of how the thing moves.

0:01:14.717,0:01:17.944
And the second, was something [br]one of our teachers told us.

0:01:17.944,0:01:20.347
He actually did the weasel in Ice Age.

0:01:20.347,0:01:21.738
And he said:

0:01:21.738,0:01:24.895
"As an animator you are not a director, you're an actor."

0:01:24.895,0:01:27.876
So, if you want to find the right motion for a character,

0:01:27.876,0:01:30.227
don't think about it, go use your body to find it --

0:01:30.227,0:01:31.863
stand in front of a mirror, act it out

0:01:31.863,0:01:33.565
in front of a camera -- whatever you need.

0:01:33.565,0:01:36.459
And then put it back in your character.

0:01:36.459,0:01:38.619
A year later I found myself at MIT

0:01:38.619,0:01:40.769
in the robotic life group, it was one of the first groups

0:01:40.769,0:01:43.414
researching the relationships [br]between humans and robots.

0:01:43.414,0:01:45.347
And I still had this dream to make

0:01:45.347,0:01:48.064
an actual, physical Luxo Jr. lamp.

0:01:48.064,0:01:49.631
But I found that robots didn't move at all

0:01:49.631,0:01:50.909
in this engaging way that I was used to

0:01:50.909,0:01:52.544
for my animation studies.

0:01:52.544,0:01:54.761
Instead, they were all --

0:01:54.761,0:01:57.255
how should I put it, they were all kind of robotic.

0:01:57.255,0:01:59.152
(Laughter)

0:01:59.152,0:02:02.787
And I thought what if I took whatever [br]I learned in animation school,

0:02:02.787,0:02:05.455
and used that to design my robotic desk lamp.

0:02:05.455,0:02:07.539
So I went and designed frame-by-frame

0:02:07.539,0:02:09.393
to try to make this robot

0:02:09.393,0:02:12.050
as graceful and engaging as possible.

0:02:12.050,0:02:14.092
And here when you see the robot interacting with me

0:02:14.092,0:02:15.723
on a desktop.

0:02:15.723,0:02:17.934
And I'm actually really [unclear] the robot so,

0:02:17.934,0:02:19.731
unbenounced to itself,

0:02:19.731,0:02:22.419
it's kind of digging its own grave by helping me.

0:02:22.419,0:02:24.449
(Laughter)

0:02:24.449,0:02:26.418
I wanted it to be less of a mechanical structure

0:02:26.418,0:02:27.712
giving me light,

0:02:27.712,0:02:30.754
and more of a helpful, kind of quiet apprentice

0:02:30.754,0:02:33.948
that's always there when you need [br]it and doesn't really interfere.

0:02:33.948,0:02:35.625
And when, for example, I'm looking for a battery

0:02:35.625,0:02:37.158
that I can't find,

0:02:37.158,0:02:41.753
in a subtle way, it will show me where the battery is.

0:02:41.753,0:02:44.319
So you can see my confusion here.

0:02:44.319,0:02:48.578
I'm not an actor.

0:02:48.578,0:02:50.026
And I want you to notice how the same

0:02:50.026,0:02:52.000
mechanical structure can at one point,

0:02:52.000,0:02:55.004
just by the way it moves seem gentle and caring --

0:02:55.004,0:02:58.237
and the other case, seem violent and confrontational.

0:02:58.237,0:03:01.548
And it's the same structure, [br]just the motion is different.

0:03:07.026,0:03:12.900
Actor: "You want to know something? [br]Well, you want to know something?

0:03:12.900,0:03:13.837
He was already dead!

0:03:13.837,0:03:17.854
Just lay there, eyes glazed over!"

0:03:17.854,0:03:18.857
(Laughter)

0:03:18.857,0:03:22.755
But, moving in graceful ways is just one [br]building block of this whole structure

0:03:22.755,0:03:24.339
called human-robot interaction.

0:03:24.339,0:03:25.828
I was at the time doing my PHD,

0:03:25.828,0:03:27.998
I was working on human robot teamwork;

0:03:27.998,0:03:30.000
teams of humans and robots working together.

0:03:30.000,0:03:31.334
I was studying the engineering,

0:03:31.334,0:03:34.116
the psychology, the philosophy of teamwork.

0:03:34.116,0:03:35.583
And at the same time I found myself

0:03:35.583,0:03:37.438
in my own kind of teamwork situation

0:03:37.438,0:03:39.659
with a good friend of mine who is actually here.

0:03:39.659,0:03:42.273
And in that situation we can easily imagine robots

0:03:42.273,0:03:44.366
in the near future being there with us.

0:03:44.366,0:03:46.213
It was after Passover Seder.

0:03:46.213,0:03:48.053
We were folding up a lot of folding chairs,

0:03:48.053,0:03:50.986
and I was amazed at how quickly [br]we found our own rhythm.

0:03:50.986,0:03:52.553
Everybody did their own part.

0:03:52.553,0:03:54.182
We didn't have to divide our tasks.

0:03:54.182,0:03:56.278
We didn't have to communicate verbally about this.

0:03:56.278,0:03:57.889
It all just happened.

0:03:57.889,0:03:58.783
And I thought,

0:03:58.783,0:04:00.632
humans and robots don't look at all like this.

0:04:00.632,0:04:02.297
When humans and robots interact,

0:04:02.297,0:04:03.495
it's much more like a chess game.

0:04:03.495,0:04:04.758
The human does a thing,

0:04:04.758,0:04:06.638
the robot analyzes whatever the human did,

0:04:06.638,0:04:08.168
then the robot decides what to do next,

0:04:08.168,0:04:09.344
plans it and does it.

0:04:09.344,0:04:11.475
And then the human waits, until its their turn again.

0:04:11.475,0:04:12.665
So, it's much more like a chess game

0:04:12.665,0:04:14.583
and that makes sense because chess is great

0:04:14.583,0:04:16.396
for mathematicians and computer scientists.

0:04:16.396,0:04:18.654
It's all about information analysis,

0:04:18.654,0:04:21.503
decision making and planning.

0:04:21.503,0:04:25.191
But I wanted my robots to be less of a chess player,

0:04:25.191,0:04:26.860
and more like -- you know -- like a doer

0:04:26.860,0:04:29.177
that just clicks and works together.

0:04:29.177,0:04:32.570
So I made my second horrible career choice:

0:04:32.570,0:04:35.230
I decided to study acting for a semester.

0:04:35.230,0:04:38.090
I took off from a PHD, I went to acting classes.

0:04:38.090,0:04:40.043
I actually participated in a play,

0:04:40.043,0:04:43.342
I hope theres no video of that around still. (Laughter)

0:04:43.342,0:04:45.969
And I got every book I could find about acting.

0:04:45.969,0:04:47.945
Including one from the 19th century

0:04:47.945,0:04:49.012
that I got from the library.

0:04:49.012,0:04:52.470
And I was really amazed because [br]my name was the second on the list --

0:04:52.470,0:04:55.286
the previous name was in 1889. (Laughter)

0:04:55.286,0:04:57.000
And this book was kind of waiting for 100 years

0:04:57.000,0:05:00.191
to be rediscovered for robotics.

0:05:00.191,0:05:01.669
And this book shows actors

0:05:01.669,0:05:04.034
how to move every muscle in the body

0:05:04.034,0:05:06.742
to match every kind of emotion [br]that they want to express.

0:05:06.742,0:05:08.574
But the real relevation was

0:05:08.574,0:05:09.910
when I learned about method acting.

0:05:09.910,0:05:12.186
It became very popular in the 20th century.

0:05:12.186,0:05:14.830
And method acting said, you don't have [br]to plan every muscle in your body.

0:05:14.830,0:05:17.503
Instead you have to use your body [br]to find the right movement.

0:05:17.503,0:05:19.683
You have to use your sense memory

0:05:19.683,0:05:21.539
to reconstruct the emotions and kind of

0:05:21.539,0:05:24.440
think with your body to find the right expression.

0:05:24.440,0:05:26.460
Improvise, play off yor scene partner.

0:05:26.460,0:05:29.708
And this came at the same time [br]as I was reading about this trend

0:05:29.708,0:05:32.804
in cognitive psychology called embodied cognition.

0:05:32.804,0:05:34.878
Which also talks about the same ideas --

0:05:34.878,0:05:35.787
We use our bodies to think,

0:05:35.787,0:05:38.212
we don't just think with our brains [br]and use our bodies to move.

0:05:38.212,0:05:40.621
but our bodies feed back into our brain

0:05:40.621,0:05:42.962
to generate the way that we behave.

0:05:42.962,0:05:44.117
And it was like a lightning bolt.

0:05:44.117,0:05:45.356
I went back to my office.

0:05:45.356,0:05:48.840
I wrote this paper -- which I never really published

0:05:48.840,0:05:50.836
called "Acting Lessons for Artificial Intelligence."

0:05:50.836,0:05:52.233
And I even took another month

0:05:52.233,0:05:54.844
to do what was then the first theater play

0:05:54.844,0:05:56.696
with a human and a robot acting together.

0:05:56.696,0:06:00.494
That's what you say before with an actress.

0:06:00.494,0:06:01.692
And I thought:

0:06:01.692,0:06:04.599
How can we make an artificial intelligence model --

0:06:04.599,0:06:06.460
computer, computational model --

0:06:06.460,0:06:08.784
that will model some of these ideas of improvisation?

0:06:08.784,0:06:10.941
Of taking risks of taking chances.

0:06:10.941,0:06:12.508
Even of making mistakes.

0:06:12.508,0:06:15.280
Maybe it can make for better robotic teammates.

0:06:15.280,0:06:17.908
So I worked for quite a long time on these models

0:06:17.908,0:06:20.257
and I implemented them on a number of robots.

0:06:20.257,0:06:22.272
Here you can see a very early example

0:06:22.272,0:06:25.984
with the robots trying to use this [br]embodied artificial intelligence,

0:06:25.984,0:06:28.742
to try to match my movements as closely as possible.

0:06:28.742,0:06:30.229
It's sort of like a game.

0:06:30.229,0:06:32.086
Let's look at it.

0:06:35.594,0:06:39.698
You can see when I psych it out, it gets fooled.

0:06:39.698,0:06:41.899
And it's a little bit like what you might see actors do

0:06:41.899,0:06:43.786
when they try to mirror each other

0:06:43.786,0:06:46.304
to find the right synchrony between them.

0:06:46.304,0:06:48.220
And then, I did another experiment,

0:06:48.220,0:06:52.303
and I got people off the street [br]to use the robotic desk lamp,

0:06:52.303,0:06:55.920
and try out this idea of embodied artificial intelligence.

0:06:55.920,0:06:57.177
So,

0:06:57.177,0:07:00.451
I actually used two kinds of brains for the same robot.

0:07:00.451,0:07:01.863
The robot is the same lamp that you saw,

0:07:01.863,0:07:03.793
and I put in it two brains.

0:07:03.793,0:07:05.633
For one half of the people,

0:07:05.633,0:07:08.457
I put in a brain that's kind of the traditional,

0:07:08.457,0:07:09.711
calculated robotic brain.

0:07:09.711,0:07:12.650
It waits for its turn, it analyzes everything, it plans --

0:07:12.650,0:07:13.921
let's call is the call it the calculated brain.

0:07:13.921,0:07:17.553
The other, got more the stage actor, risk taker brain.

0:07:17.553,0:07:19.633
Let's call it the adventurous brain.

0:07:19.633,0:07:22.594
It sometimes acts without knowing [br]everything it has to know.

0:07:22.594,0:07:24.771
It sometimes makes mistakes and corrects them.

0:07:24.771,0:07:27.354
And I had them do this very tedious task

0:07:27.354,0:07:28.852
that took almost 20 minutes

0:07:28.852,0:07:30.353
and they had to work together.

0:07:30.353,0:07:32.634
Somehow simulating like a factory job

0:07:32.634,0:07:35.299
of repetitively doing the same thing.

0:07:35.299,0:07:37.073
And what I found was that people actually loved

0:07:37.073,0:07:38.721
the adventurous robot.

0:07:38.721,0:07:40.103
And they thought it was more intelligent,

0:07:40.103,0:07:42.171
more committed, a better member of the team,

0:07:42.171,0:07:44.268
contributed to the success of the team more.

0:07:44.268,0:07:45.983
They even called it 'he' and 'she',

0:07:45.983,0:07:48.899
whereas people with the calculated brain called it 'it'.

0:07:48.899,0:07:51.646
And nobody ever called it 'he' or 'she'.

0:07:51.646,0:07:53.495
When they talked about it after the task

0:07:53.495,0:07:55.193
with adventurous brain, (laughter)

0:07:55.193,0:07:59.275
they said,"by the end we were good [br]friends and high-fived mentally."

0:07:59.275,0:08:00.714
Whatever that means.

0:08:00.714,0:08:03.961
(Laughter)

0:08:03.961,0:08:06.676
Whereas the people with the calculated brain

0:08:06.676,0:08:09.049
said it was just like a lazy apprentice.

0:08:09.049,0:08:11.750
It only did what it was supposed [br]to do and nothing more.

0:08:11.750,0:08:14.307
Which is almost what people expect robots to do,

0:08:14.307,0:08:16.742
so I was surprised that people [br]had higher expectations

0:08:16.742,0:08:21.886
of robots, than what anybody in robotics [br]thought robots should be doing.

0:08:21.886,0:08:23.856
And in a way, I thought, maybe it's time --

0:08:23.856,0:08:26.717
just like method acting changed the way

0:08:26.717,0:08:28.149
people thought about acting in the 19th century,

0:08:28.149,0:08:29.992
from going from the very calculated,

0:08:29.992,0:08:31.774
planned way of behaving,

0:08:31.774,0:08:35.295
to a more intuitive, risk-taking, [br]embodied way of behaving.

0:08:35.295,0:08:36.699
Maybe it's time for robots

0:08:36.699,0:08:39.845
to have the same kind of revolution.

0:08:39.845,0:08:40.966
A few years later,

0:08:40.966,0:08:43.474
I was at my next research job [br]at Georgia Tech in Atlanta,

0:08:43.474,0:08:44.539
and I was working in a group

0:08:44.539,0:08:45.821
dealing with robotic musicians.

0:08:45.821,0:08:48.662
And I thought -- music -- that's the perfect place

0:08:48.662,0:08:51.059
to look at teamwork, coordination,

0:08:51.059,0:08:52.948
timing, improvisation --

0:08:52.948,0:08:55.484
and we just got this robot playing marimba.

0:08:55.484,0:08:57.499
Marimba, for everybody [unclear] like me,

0:08:57.499,0:09:00.344
it was this huge, wooden xylophone.

0:09:00.344,0:09:03.039
And, when I was looking at this,

0:09:03.039,0:09:05.633
I looked at other works in [br]human-robot improvisation.

0:09:05.633,0:09:08.012
Yes, there are other works in [br]human-robot improvisation,

0:09:08.012,0:09:10.105
and they were also a little bit like a chess game.

0:09:10.105,0:09:11.409
The human would play,

0:09:11.409,0:09:13.989
the robot would analyze what was played --

0:09:13.989,0:09:16.201
would improvise their own part.

0:09:16.201,0:09:17.868
So, this is what musicians called

0:09:17.868,0:09:19.351
a call and response interaction,

0:09:19.351,0:09:23.188
and it also fits very well, robots [br]and artificial intelligence.

0:09:23.188,0:09:25.680
But I thought, if I use the same ideas I used in the

0:09:25.680,0:09:28.232
theater play and in the teamwork studies,

0:09:28.232,0:09:30.992
maybe I can make the robots jam together

0:09:30.992,0:09:32.305
like a band.

0:09:32.305,0:09:36.118
Everybody's riffing off each other, [br]nobody is stopping it for a moment.

0:09:36.118,0:09:38.910
And so, I tried to do the same [br]things, this time with music,

0:09:38.910,0:09:40.297
where the robot doesn't really know

0:09:40.297,0:09:41.272
what it's about to play.

0:09:41.272,0:09:42.545
It just sort of moves its body

0:09:42.545,0:09:44.773
and uses opportunities to play,

0:09:44.773,0:09:47.494
And does what my chess teacher, [br]when I was 17, taught me.

0:09:47.494,0:09:48.758
She said, when you improvise,

0:09:48.758,0:09:50.419
sometimes you don't know what you're doing

0:09:50.419,0:09:51.434
and you're still doing it.

0:09:51.434,0:09:52.864
And so I tried to make a robot that doesn't actually

0:09:52.864,0:09:54.809
know what it's doing, but it's still doing it.

0:09:54.809,0:09:57.753
So let's look at a few seconds [br]from this performance.

0:09:57.753,0:10:00.670
Where the robot listens to the human musician

0:10:00.670,0:10:02.473
and improvises.

0:10:02.473,0:10:04.846
And then, look how the human musician also

0:10:04.846,0:10:06.633
responds to what the robot is doing, and picking up

0:10:06.633,0:10:09.453
from its behavior.

0:10:09.453,0:10:13.981
And at some point can even be surprised [br]by what the robot came up with.

0:10:13.981,0:11:00.056
(Music is played by the musician and the robot)

0:11:00.056,0:11:05.380
(Applause)

0:11:05.380,0:11:07.073
Being a musician is not just about making notes,

0:11:07.073,0:11:09.459
otherwise nobody would every go see a live show.

0:11:09.459,0:11:11.382
Musicians also communicate with their bodies,

0:11:11.382,0:11:13.329
with other band members, with the audience,

0:11:13.329,0:11:15.492
they use their bodies to express the music.

0:11:15.492,0:11:17.872
And I thought, we already have [br]a robot musician on stage,

0:11:17.872,0:11:20.820
why not make it be a full-fledged musician.

0:11:20.820,0:11:23.336
And I started designing a socially expressive head

0:11:23.336,0:11:24.885
for the robot.

0:11:24.885,0:11:26.712
The head does't actually touch the marimba,

0:11:26.712,0:11:28.348
it just expresses what the music is like.

0:11:28.348,0:11:31.428
These are some napkin sketches [br]from a bar in Atlanta,

0:11:31.428,0:11:34.050
that was dangerously located exactly halfway

0:11:34.050,0:11:35.502
between my lab and my home. (Laughter)

0:11:35.502,0:11:36.918
So I spent, I would say on average,

0:11:36.918,0:11:39.846
three to four hours a day there.

0:11:39.846,0:11:42.744
I think. (Laughter)

0:11:42.744,0:11:45.723
And I went back to my animation [br]tools and tried to figure out

0:11:45.723,0:11:47.905
not just what a robotic musician would look like,

0:11:47.905,0:11:50.713
but especially what a robotic [br]musician would move like.

0:11:50.713,0:11:54.013
To sort of show that it doesn't like [br]what the other person is playing --

0:11:54.013,0:11:56.492
and maybe show whatever beat it's feeling

0:11:56.492,0:11:58.141
at the moment. (Laughter)

0:11:58.141,0:12:02.668
So we ended up actually getting the money [br]to build this robot which was nice.

0:12:02.668,0:12:04.648
I'm going to show you now the [br]same kind of performance --

0:12:04.648,0:12:07.220
this time with a socially expressive head.

0:12:07.220,0:12:09.184
And notice one thing --

0:12:09.184,0:12:10.880
how the robot is really showing us

0:12:10.880,0:12:12.744
the beat it's picking up from the human.

0:12:12.744,0:12:16.760
When also giving the human a sense, [br]like the robot knows what it's doing.

0:12:16.760,0:12:18.140
And also how it changes the way it moves[br]

0:12:18.140,0:12:21.820
as soon as it starts its own solo.

0:12:21.820,0:12:49.064
(Robot interacting with music played by musician)

0:12:49.064,0:12:52.228
And now look at the final chord of the piece again,

0:12:52.228,0:12:55.067
and this time the robot communicates with its body

0:12:55.067,0:12:57.204
when it's busy doing its own thing.

0:12:57.204,0:12:58.816
And when its ready

0:12:58.816,0:13:02.168
to coordinate the final chord with me.

0:13:02.168,0:13:14.330
(Musician and robot playing music together)

0:13:14.330,0:13:17.890
(Robot and musician play final chord in synchrony)

0:13:17.890,0:13:20.906
(Applause)

0:13:20.906,0:13:25.263
Thanks. I hope you see how much this,

0:13:25.263,0:13:27.944
how much this part of the body [br]that doesn't touch the instrument

0:13:27.944,0:13:32.141
actually helps with the musical performance.

0:13:32.141,0:13:34.530
We are in Atlanta, so obviously some rapper

0:13:34.530,0:13:36.399
will come into our lab at some point. (Laughter)

0:13:36.399,0:13:38.639
And we had this rapper come in

0:13:38.639,0:13:41.212
and do a little jam with the robot.

0:13:41.212,0:13:43.914
And here you can see the robot

0:13:43.914,0:13:45.143
basically responding to the beat and --

0:13:45.143,0:13:47.997
notice two things. One, how irresistible it is

0:13:47.997,0:13:50.510
to join the robot while it's moving its head.

0:13:50.051,0:13:52.028
and you kind of want to move [br]your own head when it does it.

0:13:52.028,0:13:56.207
And second, even though the rapper [br]is really focused on his iPhone,

0:13:56.207,0:13:59.121
as soon as the robot turns to him, he turns back.

0:13:59.121,0:14:01.200
So even though it's just in [br]the periphery of his vision --

0:14:01.200,0:14:03.770
it's just in the corner of his eye -- [br]it's very powerful.

0:14:03.770,0:14:05.821
And the reason is that we can't ignore

0:14:05.821,0:14:07.522
physical things moving our environment.

0:14:07.522,0:14:09.286
We are wired for that.

0:14:09.286,0:14:12.618
So, if you have a problem with maybe your partners

0:14:12.618,0:14:15.813
looking at the iPhone too much [br]or their smartphone too much,

0:14:15.813,0:14:16.852
you might want to have a robot there

0:14:16.852,0:14:18.636
to get their attention. (Laughter)

0:14:18.636,0:14:36.886
(Rapping to music)

0:14:36.886,0:14:39.011
(Laughter)

0:14:39.011,0:14:45.611
(Applause)

0:14:45.611,0:14:48.167
Just to introduce the last robot

0:14:48.167,0:14:49.742
that we've worked on --

0:14:49.742,0:14:51.353
that came out of something kind [br]of surprising that we found --

0:14:51.353,0:14:54.482
at some point people didn't care anymore [br]about the robot being so intelligent,

0:14:54.482,0:14:56.035
and can improvise and listen,

0:14:56.035,0:15:00.750
and do all these embodied intelligence [br]things that I spent years on developing.

0:15:00.750,0:15:04.263
They really like that the robot [br]was enjoying the music.

0:15:04.263,0:15:06.315
And they didn't say that the [br]robot was moving to the music,

0:15:06.315,0:15:08.340
they said that the robot was enjoying the music.

0:15:08.340,0:15:10.744
And we thought, why don't we take this idea,

0:15:10.744,0:15:13.629
and I designed a new piece of furniture.

0:15:13.629,0:15:15.718
This time it wasn't a desk [br]lamp it was a speaker dock.

0:15:15.718,0:15:18.826
It was one of those things you [br]plug your smartphone in.

0:15:18.826,0:15:20.545
And I thought, what would happen

0:15:20.545,0:15:23.146
if your speaker dock didn't [br]just play the music for you,

0:15:23.146,0:15:25.657
but it would actually enjoy it to. (Laughter)

0:15:25.657,0:15:27.434
And so again, here are some animation tests

0:15:27.434,0:15:31.835
from an early stage. (Laughter)

0:15:31.835,0:15:36.005
And this is what the final product looked like.

0:15:47.223,0:16:09.461
(Music playing in synchrony [br]with the robot's movements)

0:16:09.461,0:16:12.105
So, a lot of bobbing head.

0:16:12.105,0:16:15.697
(Applause)

0:16:15.697,0:16:17.228
A lot of bobbing heads in the audience,

0:16:17.228,0:16:20.386
so we can still see robots influence people.

0:16:20.386,0:16:23.028
And it's not just fun and games.

0:16:23.028,0:16:25.128
I think one of the reasons I care so much

0:16:25.128,0:16:27.214
about robots that use their body to communicate

0:16:27.214,0:16:29.407
and use their body to move --

0:16:29.407,0:16:32.640
and I'm going to let you in on a little [br]secret we roboticists are hiding --

0:16:32.640,0:16:34.632
is that every one of you is [br]going to be living with a robot

0:16:34.632,0:16:37.254
at some point in their life. (Laughter)

0:16:37.254,0:16:39.550
Somewhere in your future there's [br]going to be a robot in your life.

0:16:39.550,0:16:41.678
And if not in yours, then in your children's life.

0:16:41.678,0:16:43.467
And I want these robots to be --

0:16:43.467,0:16:46.961
to be more fluent, more engaging, more graceful

0:16:46.961,0:16:48.749
than currently they seem to be.

0:16:48.749,0:16:50.668
And for that I think that maybe robots

0:16:50.668,0:16:52.034
need to be less like chess players

0:16:52.034,0:16:54.744
and more like stage actors and more like musicians.

0:16:54.744,0:16:57.683
Maybe they should be able to [br]take chances and improvise.

0:16:57.683,0:17:00.180
And maybe they should be able to [br]anticipate what you're about to do.

0:17:00.180,0:17:02.756
And maybe they need to be able to make mistakes

0:17:02.756,0:17:03.979
and correct them,

0:17:03.979,0:17:06.015
because in the end we are human.

0:17:06.015,0:17:09.386
And maybe as humans, robots [br]that are a little less than perfect

0:17:09.386,0:17:11.252
are just perfect for us.

0:17:11.252,0:17:12.766
Thank you.

0:17:12.766,0:17:16.188
(Applause)