TEDxJaffa 2013
Guy Hoffman: Robots with "soul"
Filmed:
Readability: 3.8
3,054,391 views
What kind of robots does an animator / jazz musician / roboticist make? Playful, reactive, curious ones. Guy Hoffman shows demo film of his family of unusual robots -- including two musical bots that like to jam with humans.
Guy Hoffman - Roboticist
Can robots and humans interact the way that human beings interact with each other? Guy Hoffman researches embodied cognition and intelligence in robots. Full bio
Can robots and humans interact the way that human beings interact with each other? Guy Hoffman researches embodied cognition and intelligence in robots. Full bio
Double-click the English transcript below to play the video.
00:12
My job is to design, build and study
0
564
2522
00:15
robots that communicate with people.
1
3086
2000
00:17
But this story doesn't start with robotics at all,
2
5086
1553
00:18
it starts with animation.
3
6639
1892
00:20
When I first saw Pixar's "Luxo Jr.,"
4
8531
2586
00:23
I was amazed by how much emotion
5
11117
1986
00:25
they could put into something
6
13103
1758
00:27
as trivial as a desk lamp.
7
14861
2591
00:29
I mean, look at them -- at the end of this movie,
8
17452
1833
00:31
you actually feel something for two pieces of furniture.
9
19285
2504
00:33
(Laughter)
10
21789
1828
00:35
And I said, I have to learn how to do this.
11
23617
2120
00:37
So I made a really bad career decision.
12
25737
3534
00:41
And that's what my mom was like when I did it.
13
29271
2404
00:43
(Laughter)
14
31675
2011
00:45
I left a very cozy tech job in Israel
15
33686
2180
00:48
at a nice software company and I moved to New York
16
35866
2605
00:50
to study animation.
17
38471
1011
00:51
And there I lived
18
39482
1157
00:52
in a collapsing apartment building
in Harlem with roommates.
in Harlem with roommates.
19
40639
2996
00:55
I'm not using this phrase metaphorically,
20
43635
1368
00:57
the ceiling actually collapsed one day
21
45003
1960
00:59
in our living room.
22
46963
1310
01:00
Whenever they did those news stories
about building violations in New York,
about building violations in New York,
23
48273
2666
01:03
they would put the report in front of our building.
24
50939
2246
01:05
As kind of like a backdrop
to show how bad things are.
to show how bad things are.
25
53185
3598
01:08
Anyway, during the day I went to school and at night
26
56783
1946
01:10
I would sit and draw frame by frame
of pencil animation.
of pencil animation.
27
58729
3522
01:14
And I learned two surprising lessons --
28
62251
2274
01:16
one of them was that
29
64525
2553
01:19
when you want to arouse emotions,
30
67078
1960
01:21
it doesn't matter so much how something looks,
31
69038
1726
01:22
it's all in the motion -- it's in the timing
32
70764
2346
01:25
of how the thing moves.
33
73110
1607
01:26
And the second, was something
one of our teachers told us.
one of our teachers told us.
34
74717
3227
01:30
He actually did the weasel in Ice Age.
35
77944
2403
01:32
And he said:
36
80347
1391
01:33
"As an animator you are not
a director, you're an actor."
a director, you're an actor."
37
81738
3157
01:37
So, if you want to find the
right motion for a character,
right motion for a character,
38
84895
2981
01:40
don't think about it, go use your body to find it --
39
87876
2351
01:42
stand in front of a mirror, act it out
40
90227
1636
01:44
in front of a camera -- whatever you need.
41
91863
1702
01:45
And then put it back in your character.
42
93565
2894
01:48
A year later I found myself at MIT
43
96459
2160
01:50
in the robotic life group, it was one of the first groups
44
98619
2150
01:52
researching the relationships
between humans and robots.
between humans and robots.
45
100769
2645
01:55
And I still had this dream to make
46
103414
1933
01:57
an actual, physical Luxo Jr. lamp.
47
105347
2717
02:00
But I found that robots didn't move at all
48
108064
1567
02:01
in this engaging way that I was used to
49
109631
1278
02:03
for my animation studies.
50
110909
1635
02:04
Instead, they were all --
51
112544
2217
02:06
how should I put it, they were all kind of robotic.
52
114761
2494
02:09
(Laughter)
53
117255
1897
02:11
And I thought, what if I took whatever
I learned in animation school,
I learned in animation school,
54
119152
3635
02:14
and used that to design my robotic desk lamp.
55
122787
2668
02:17
So I went and designed frame by frame
56
125455
2084
02:19
to try to make this robot
57
127539
1854
02:21
as graceful and engaging as possible.
58
129393
2657
02:24
And here when you see the robot interacting with me
59
132050
2042
02:26
on a desktop.
60
134092
1631
02:27
And I'm actually redesigning the robot so,
61
135723
2211
02:30
unbeknownst to itself,
62
137934
1797
02:31
it's kind of digging its own grave by helping me.
63
139731
2688
02:34
(Laughter)
64
142419
2030
02:36
I wanted it to be less of a mechanical structure
65
144449
1969
02:38
giving me light,
66
146418
1294
02:39
and more of a helpful, kind of quiet apprentice
67
147712
3042
02:42
that's always there when you need
it and doesn't really interfere.
it and doesn't really interfere.
68
150754
3194
02:46
And when, for example, I'm looking for a battery
69
153948
1677
02:47
that I can't find,
70
155625
1533
02:49
in a subtle way, it will show me where the battery is.
71
157158
4595
02:53
So you can see my confusion here.
72
161753
2566
02:56
I'm not an actor.
73
164319
4259
03:00
And I want you to notice how the same
74
168578
1448
03:02
mechanical structure can at one point,
75
170026
1974
03:04
just by the way it moves seem gentle and caring --
76
172000
3004
03:07
and in the other case, seem
violent and confrontational.
violent and confrontational.
77
175004
3233
03:10
And it's the same structure,
just the motion is different.
just the motion is different.
78
178237
3311
03:19
Actor: "You want to know something?
Well, you want to know something?
Well, you want to know something?
79
187026
5874
03:25
He was already dead!
80
192900
937
03:26
Just laying there, eyes glazed over!"
81
193837
4017
03:30
(Laughter)
82
197854
1003
03:31
But, moving in graceful ways is just one
building block of this whole structure
building block of this whole structure
83
198857
3898
03:34
called human-robot interaction.
84
202755
1584
03:36
I was at the time doing my Ph.D.,
85
204339
1489
03:38
I was working on human robot teamwork;
86
205828
2170
03:40
teams of humans and robots working together.
87
207998
2002
03:42
I was studying the engineering,
88
210000
1334
03:43
the psychology, the philosophy of teamwork.
89
211334
2782
03:46
And at the same time I found myself
90
214116
1467
03:47
in my own kind of teamwork situation
91
215583
1855
03:49
with a good friend of mine who is actually here.
92
217438
2221
03:51
And in that situation we can easily imagine robots
93
219659
2614
03:54
in the near future being there with us.
94
222273
2093
03:56
It was after a Passover seder.
95
224366
1847
03:58
We were folding up a lot of folding chairs,
96
226213
1840
04:00
and I was amazed at how quickly
we found our own rhythm.
we found our own rhythm.
97
228053
2933
04:03
Everybody did their own part.
98
230986
1567
04:04
We didn't have to divide our tasks.
99
232553
1629
04:06
We didn't have to communicate verbally about this.
100
234182
2096
04:08
It all just happened.
101
236278
1611
04:10
And I thought,
102
237889
894
04:10
humans and robots don't look at all like this.
103
238783
1849
04:12
When humans and robots interact,
104
240632
1665
04:14
it's much more like a chess game.
105
242297
1198
04:15
The human does a thing,
106
243495
1263
04:16
the robot analyzes whatever the human did,
107
244758
1880
04:18
then the robot decides what to do next,
108
246638
1530
04:20
plans it and does it.
109
248168
1176
04:21
And then the human waits, until it's their turn again.
110
249344
2131
04:23
So, it's much more like a chess game
111
251475
1190
04:24
and that makes sense because chess is great
112
252665
2210
04:27
for mathematicians and computer scientists.
113
254875
1521
04:28
It's all about information analysis,
114
256396
2258
04:30
decision making and planning.
115
258654
2849
04:33
But I wanted my robot to be less of a chess player,
116
261503
3688
04:37
and more like a doer
117
265191
1961
04:39
that just clicks and works together.
118
267152
2025
04:41
So I made my second horrible career choice:
119
269177
3393
04:44
I decided to study acting for a semester.
120
272570
2660
04:47
I took off from a Ph.D. I went to acting classes.
121
275230
2860
04:50
I actually participated in a play,
122
278090
2460
04:52
I hope theres no video of that around still.
123
280550
2792
04:55
And I got every book I could find about acting,
124
283342
2627
04:58
including one from the 19th century
125
285969
1976
05:00
that I got from the library.
126
287945
1067
05:01
And I was really amazed because my
name was the second name on the list --
name was the second name on the list --
127
289012
3458
05:04
the previous name was in 1889. (Laughter)
128
292470
2816
05:07
And this book was kind of waiting for 100 years
129
295286
1714
05:09
to be rediscovered for robotics.
130
297000
3191
05:12
And this book shows actors
131
300191
1478
05:13
how to move every muscle in the body
132
301669
2365
05:16
to match every kind of emotion
that they want to express.
that they want to express.
133
304034
2708
05:18
But the real revelation was
134
306742
1832
05:20
when I learned about method acting.
135
308574
1336
05:22
It became very popular in the 20th century.
136
309910
2276
05:24
And method acting said, you don't have
to plan every muscle in your body.
to plan every muscle in your body.
137
312186
2644
05:27
Instead you have to use your body
to find the right movement.
to find the right movement.
138
314830
2673
05:29
You have to use your sense memory
139
317503
2180
05:31
to reconstruct the emotions and kind of
140
319683
1856
05:33
think with your body to find the right expression.
141
321539
2901
05:36
Improvise, play off yor scene partner.
142
324440
2020
05:38
And this came at the same time
as I was reading about this trend
as I was reading about this trend
143
326460
3248
05:41
in cognitive psychology called embodied cognition.
144
329708
3096
05:45
Which also talks about the same ideas --
145
332804
1674
05:46
We use our bodies to think,
146
334478
1509
05:48
we don't just think with our brains
and use our bodies to move.
and use our bodies to move.
147
335987
2225
05:50
but our bodies feed back into our brain
148
338212
2409
05:52
to generate the way that we behave.
149
340621
2217
05:55
And it was like a lightning bolt.
150
342838
1371
05:56
I went back to my office.
151
344209
1362
05:57
I wrote this paper -- which I never really published
152
345571
2899
06:00
called "Acting Lessons for Artificial Intelligence."
153
348470
2366
06:03
And I even took another month
154
350836
1397
06:04
to do what was then the first theater play
155
352233
2611
06:07
with a human and a robot acting together.
156
354844
1852
06:08
That's what you saw before with the actors.
157
356696
3567
06:12
And I thought:
158
360263
1429
06:13
How can we make an artificial intelligence model --
159
361692
2907
06:16
computer, computational model --
160
364599
1861
06:18
that will model some of these ideas of improvisation,
161
366460
2324
06:20
of taking risks, of taking chances,
162
368784
2157
06:23
even of making mistakes.
163
370941
1567
06:24
Maybe it can make for better robotic teammates.
164
372508
2772
06:27
So I worked for quite a long time on these models
165
375280
2628
06:30
and I implemented them on a number of robots.
166
377908
2349
06:32
Here you can see a very early example
167
380257
2015
06:34
with the robots trying to use this
embodied artificial intelligence,
embodied artificial intelligence,
168
382272
3712
06:38
to try to match my movements
as closely as possible,
as closely as possible,
169
385984
2758
06:40
sort of like a game.
170
388742
1487
06:42
Let's look at it.
171
390229
1857
06:47
You can see when I psych it out, it gets fooled.
172
395594
4104
06:51
And it's a little bit like what you might see actors do
173
399698
2201
06:54
when they try to mirror each other
174
401899
1887
06:55
to find the right synchrony between them.
175
403786
2518
06:58
And then, I did another experiment,
176
406304
1916
07:00
and I got people off the street
to use the robotic desk lamp,
to use the robotic desk lamp,
177
408220
4083
07:04
and try out this idea of embodied
artificial intelligence.
artificial intelligence.
178
412303
3954
07:08
So, I actually used two kinds
of brains for the same robot.
of brains for the same robot.
179
416257
4286
07:12
The robot is the same lamp that you saw,
180
420543
1320
07:14
and I put in it two brains.
181
421863
1930
07:15
For one half of the people,
182
423793
1840
07:17
I put in a brain that's kind of the traditional,
183
425633
2824
07:20
calculated robotic brain.
184
428457
1254
07:21
It waits for its turn, it analyzes everything, it plans.
185
429711
2554
07:24
Let's call it the calculated brain.
186
432265
1656
07:26
The other got more the stage actor, risk taker brain.
187
433921
3632
07:29
Let's call it the adventurous brain.
188
437553
2080
07:31
It sometimes acts without knowing
everything it has to know.
everything it has to know.
189
439633
2961
07:34
It sometimes makes mistakes and corrects them.
190
442594
2177
07:36
And I had them do this very tedious task
191
444771
2583
07:39
that took almost 20 minutes
192
447354
1498
07:41
and they had to work together.
193
448852
1501
07:42
Somehow simulating like a factory job
194
450353
2281
07:44
of repetitively doing the same thing.
195
452634
2665
07:47
And what I found was that people actually loved
196
455299
1774
07:49
the adventurous robot.
197
457073
1648
07:50
And they thought it was more intelligent,
198
458721
1382
07:52
more committed, a better member of the team,
199
460103
2068
07:54
contributed to the success of the team more.
200
462171
2097
07:56
They even called it 'he' and 'she,'
201
464268
1715
07:58
whereas people with the calculated brain called it 'it.'
202
465983
2916
08:01
And nobody ever called it 'he' or 'she'.
203
468899
2747
08:03
When they talked about it after the task
204
471646
1849
08:05
with the adventurous brain,
205
473495
1698
08:07
they said, "By the end, we were good
friends and high-fived mentally."
friends and high-fived mentally."
206
475193
4082
08:11
Whatever that means.
207
479275
1439
08:12
(Laughter) Sounds painful.
208
480714
3247
08:16
Whereas the people with the calculated brain
209
483961
2715
08:18
said it was just like a lazy apprentice.
210
486676
2373
08:21
It only did what it was supposed
to do and nothing more.
to do and nothing more.
211
489049
2701
08:23
Which is almost what people expect robots to do,
212
491750
2557
08:26
so I was surprised that people
had higher expectations
had higher expectations
213
494307
2435
08:28
of robots, than what anybody in robotics
thought robots should be doing.
thought robots should be doing.
214
496742
5144
08:34
And in a way, I thought, maybe it's time --
215
501886
1970
08:36
just like method acting changed the way
216
503856
2861
08:38
people thought about acting in the 19th century,
217
506717
1432
08:40
from going from the very calculated,
218
508149
1843
08:42
planned way of behaving,
219
509992
1782
08:43
to a more intuitive, risk-taking,
embodied way of behaving.
embodied way of behaving.
220
511774
3521
08:47
Maybe it's time for robots
221
515295
1404
08:48
to have the same kind of revolution.
222
516699
3146
08:52
A few years later,
223
519845
1121
08:53
I was at my next research job
at Georgia Tech in Atlanta,
at Georgia Tech in Atlanta,
224
520966
2508
08:55
and I was working in a group
225
523474
1065
08:56
dealing with robotic musicians.
226
524539
1282
08:58
And I thought, music, that's the perfect place
227
525821
2841
09:00
to look at teamwork, coordination,
228
528662
2397
09:03
timing, improvisation --
229
531059
1889
09:05
and we just got this robot playing marimba.
230
532948
2536
09:07
Marimba, for everybody who was like me,
231
535484
2015
09:09
it was this huge, wooden xylophone.
232
537499
2845
09:12
And, when I was looking at this,
233
540344
2695
09:15
I looked at other works in
human-robot improvisation --
human-robot improvisation --
234
543039
2594
09:17
yes, there are other works in
human-robot improvisation --
human-robot improvisation --
235
545633
2379
09:20
and they were also a little bit like a chess game.
236
548012
2093
09:22
The human would play,
237
550105
1304
09:23
the robot would analyze what was played,
238
551409
2580
09:26
would improvise their own part.
239
553989
2212
09:28
So, this is what musicians called
240
556201
1667
09:30
a call and response interaction,
241
557868
1483
09:31
and it also fits very well, robots
and artificial intelligence.
and artificial intelligence.
242
559351
3837
09:35
But I thought, if I use the same ideas I used
243
563188
1907
09:37
in the theater play and in the teamwork studies,
244
565095
3137
09:40
maybe I can make the robots jam together
245
568232
2452
09:42
like a band.
246
570684
1621
09:44
Everybody's riffing off each other,
nobody is stopping it for a moment.
nobody is stopping it for a moment.
247
572305
3813
09:48
And so, I tried to do the same
things, this time with music,
things, this time with music,
248
576118
2792
09:51
where the robot doesn't really know
249
578910
1387
09:52
what it's about to play.
250
580297
975
09:53
It just sort of moves its body
251
581272
1273
09:54
and uses opportunities to play,
252
582545
2228
09:56
And does what my jazz teacher
when I was 17 taught me.
when I was 17 taught me.
253
584773
2721
09:59
She said, when you improvise,
254
587494
1264
10:00
sometimes you don't know what you're doing
255
588758
1661
10:02
and you're still doing it.
256
590419
1015
10:03
And so I tried to make a robot that doesn't actually
257
591434
1430
10:05
know what it's doing, but it's still doing it.
258
592864
1945
10:07
So let's look at a few seconds
from this performance.
from this performance.
259
594809
2944
10:09
Where the robot listens to the human musician
260
597753
2917
10:12
and improvises.
261
600670
1803
10:14
And then, look at how the human musician also
262
602473
2373
10:17
responds to what the robot is doing, and picking up
263
604846
1787
10:18
from its behavior.
264
606633
2820
10:21
And at some point can even be surprised
by what the robot came up with.
by what the robot came up with.
265
609453
4528
10:26
(Music)
266
613981
46075
11:12
(Applause)
267
660056
5031
11:17
Being a musician is not just about making notes,
268
665087
1986
11:19
otherwise nobody would ever go see a live show.
269
667073
2386
11:21
Musicians also communicate with their bodies,
270
669459
1923
11:23
with other band members, with the audience,
271
671382
1947
11:25
they use their bodies to express the music.
272
673329
2163
11:27
And I thought, we already have
a robot musician on stage,
a robot musician on stage,
273
675492
2380
11:30
why not make it be a full-fledged musician.
274
677872
2948
11:33
And I started designing a socially expressive head
275
680820
2516
11:35
for the robot.
276
683336
1549
11:37
The head does't actually touch the marimba,
277
684885
1827
11:38
it just expresses what the music is like.
278
686712
1636
11:40
These are some napkin sketches
from a bar in Atlanta,
from a bar in Atlanta,
279
688348
3080
11:43
that was dangerously located exactly halfway
280
691428
2622
11:46
between my lab and my home. (Laughter)
281
694050
1452
11:47
So I spent, I would say on average,
282
695502
1416
11:49
three to four hours a day there.
283
696918
2928
11:52
I think. (Laughter)
284
699846
2898
11:54
And I went back to my animation
tools and tried to figure out
tools and tried to figure out
285
702744
2979
11:57
not just what a robotic musician would look like,
286
705723
2182
12:00
but especially what a robotic
musician would move like.
musician would move like.
287
707905
2808
12:02
To sort of show that it doesn't like
what the other person is playing --
what the other person is playing --
288
710713
3300
12:06
and maybe show whatever beat it's feeling
289
714013
2479
12:08
at the moment.
290
716492
1649
12:10
So we ended up actually getting the money
to build this robot, which was nice.
to build this robot, which was nice.
291
718141
4527
12:14
I'm going to show you now the
same kind of performance,
same kind of performance,
292
722668
2210
12:17
this time with a socially expressive head.
293
724878
2342
12:19
And notice one thing --
294
727220
1964
12:21
how the robot is really showing us
295
729184
1696
12:23
the beat it's picking up from the human.
296
730880
1864
12:24
We're also giving the human a sense
that the robot knows what it's doing.
that the robot knows what it's doing.
297
732744
4016
12:28
And also how it changes the way it moves
298
736760
1380
12:30
as soon as it starts its own solo.
299
738140
2540
12:32
(Music)
300
740680
3920
12:36
Now it's looking at me to make sure I'm listening.
301
744600
2901
12:39
(Music)
302
747501
21100
13:00
And now look at the final chord of the piece again,
303
768601
3627
13:04
and this time the robot communicates with its body
304
772228
2839
13:07
when it's busy doing its own thing.
305
775067
2137
13:09
And when it's ready
306
777204
1612
13:11
to coordinate the final chord with me.
307
778816
3352
13:14
(Music)
308
782168
12929
13:27
(Applause)
309
795128
5778
13:33
Thanks. I hope you see how much this totally not --
310
800906
4357
13:37
how much this part of the body
that doesn't touch the instrument
that doesn't touch the instrument
311
805263
2681
13:40
actually helps with the musical performance.
312
807944
3366
13:43
And at some point, we are in Atlanta,
so obviously some rapper
so obviously some rapper
313
811310
3220
13:46
will come into our lab at some point.
314
814530
1869
13:48
And we had this rapper come in
315
816399
2240
13:50
and do a little jam with the robot.
316
818639
2573
13:53
And here you can see the robot
317
821212
2702
13:56
basically responding to the beat and --
318
823914
1229
13:57
notice two things. One, how irresistible it is
319
825143
2854
14:00
to join the robot while it's moving its head.
320
827997
2513
14:02
and you kind of want to move
your own head when it does it.
your own head when it does it.
321
830051
2361
14:04
And second, even though the rapper
is really focused on his iPhone,
is really focused on his iPhone,
322
832412
3795
14:08
as soon as the robot turns to him, he turns back.
323
836207
2914
14:11
So even though it's just in
the periphery of his vision --
the periphery of his vision --
324
839121
2079
14:13
it's just in the corner of his eye --
it's very powerful.
it's very powerful.
325
841200
2570
14:15
And the reason is that we can't ignore
326
843770
2051
14:18
physical things moving in our environment.
327
845821
1701
14:19
We are wired for that.
328
847522
1764
14:21
So, if you have a problem with maybe your partners
329
849286
3332
14:24
looking at the iPhone too much
or their smartphone too much,
or their smartphone too much,
330
852618
2871
14:27
you might want to have a robot there
331
855489
1593
14:29
to get their attention. (Laughter)
332
857082
1554
14:30
(Music)
333
858636
19575
14:50
(Applause)
334
878211
6661
14:57
Just to introduce the last robot
335
884872
2510
14:59
that we've worked on,
336
887382
2144
15:01
that came out of something kind
of surprising that we found:
of surprising that we found:
337
889526
2073
15:03
At some point people didn't care anymore
about the robot being so intelligent,
about the robot being so intelligent,
338
891599
3035
15:06
and can improvise and listen,
339
894634
1708
15:08
and do all these embodied intelligence
things that I spent years on developing.
things that I spent years on developing.
340
896342
4408
15:12
They really liked that the robot was
enjoying the music. (Laughter)
enjoying the music. (Laughter)
341
900750
3513
15:16
And they didn't say that the
robot was moving to the music,
robot was moving to the music,
342
904263
2405
15:18
they said that the robot was enjoying the music.
343
906668
1672
15:20
And we thought, why don't we take this idea,
344
908340
2404
15:22
and I designed a new piece of furniture.
345
910744
2885
15:25
This time it wasn't a desk
lamp; it was a speaker dock.
lamp; it was a speaker dock.
346
913629
2089
15:27
It was one of those things you
plug your smartphone in.
plug your smartphone in.
347
915718
3108
15:31
And I thought, what would happen
348
918826
1719
15:32
if your speaker dock didn't
just play the music for you,
just play the music for you,
349
920545
2601
15:35
but it would actually enjoy it too. (Laughter)
350
923146
2511
15:37
And so again, here are some animation tests
351
925657
1777
15:39
from an early stage. (Laughter)
352
927434
4401
15:44
And this is what the final product looked like.
353
931835
4170
15:59
("Drop It Like It's Hot")
354
947223
22238
16:21
So, a lot of bobbing head.
355
969461
2644
16:24
(Applause)
356
972105
3222
16:27
A lot of bobbing heads in the audience,
357
975327
1901
16:29
so we can still see robots influence people.
358
977228
3158
16:32
And it's not just fun and games.
359
980386
2642
16:35
I think one of the reasons I care so much
360
983028
2392
16:37
about robots that use their body to communicate
361
985420
1794
16:39
and use their body to move --
362
987214
2193
16:41
and I'm going to let you in on a little
secret we roboticists are hiding --
secret we roboticists are hiding --
363
989407
3233
16:44
is that every one of you is
going to be living with a robot
going to be living with a robot
364
992640
2361
16:47
at some point in their life.
365
995001
2253
16:49
Somewhere in your future there's
going to be a robot in your life.
going to be a robot in your life.
366
997254
2296
16:51
And if not in yours, then in your children's lives.
367
999550
2128
16:53
And I want these robots to be --
368
1001678
1789
16:55
to be more fluent, more engaging, more graceful
369
1003467
3494
16:59
than currently they seem to be.
370
1006961
1788
17:00
And for that I think that maybe robots
371
1008749
1919
17:02
need to be less like chess players
372
1010668
1366
17:04
and more like stage actors and more like musicians.
373
1012034
2710
17:06
Maybe they should be able to
take chances and improvise.
take chances and improvise.
374
1014744
2939
17:09
And maybe they should be able to
anticipate what you're about to do.
anticipate what you're about to do.
375
1017683
2497
17:12
And maybe they need to be able to make mistakes
376
1020180
2576
17:14
and correct them,
377
1022756
1223
17:16
because in the end we are human.
378
1023979
2036
17:18
And maybe as humans, robots
that are a little less than perfect
that are a little less than perfect
379
1026015
3371
17:21
are just perfect for us.
380
1029386
1866
17:23
Thank you.
381
1031252
1514
17:24
(Applause)
382
1032766
3422
ABOUT THE SPEAKER
Guy Hoffman - RoboticistCan robots and humans interact the way that human beings interact with each other? Guy Hoffman researches embodied cognition and intelligence in robots.
Why you should listen
As co-director of the IDC Media Invention Lab, Guy Hoffman researches robots with soul. He explores the humanity of robots -- how they think, feel, act, and move, as they interact with humans. He and his team staged the world’s first human-robot theater piece, as well as the first human-robot jazz duet, improv and all.
Hoffman’s work has been named one of TIME Magazine’s Best Inventions of the Year, and in 2010 and 2012, he was listed as one of Israel’s most promising researchers under 40.
Guy Hoffman | Speaker | TED.com