TED2005
Ray Kurzweil: The accelerating power of technology
Filmed:
Readability: 5.3
2,876,494 views
Inventor, entrepreneur and visionary Ray Kurzweil explains in abundant, grounded detail why, by the 2020s, we will have reverse-engineered the human brain and nanobots will be operating your consciousness.
Ray Kurzweil - Inventor, futurist
Ray Kurzweil is an engineer who has radically advanced the fields of speech, text and audio technology. He's revered for his dizzying -- yet convincing -- writing on the advance of technology, the limits of biology and the future of the human species. Full bio
Ray Kurzweil is an engineer who has radically advanced the fields of speech, text and audio technology. He's revered for his dizzying -- yet convincing -- writing on the advance of technology, the limits of biology and the future of the human species. Full bio
Double-click the English transcript below to play the video.
00:24
Well, it's great to be here.
0
0
1000
00:25
We've heard a lot about the promise of technology, and the peril.
1
1000
5000
00:30
I've been quite interested in both.
2
6000
2000
00:32
If we could convert 0.03 percent
3
8000
4000
00:36
of the sunlight that falls on the earth into energy,
4
12000
2000
00:38
we could meet all of our projected needs for 2030.
5
14000
5000
00:43
We can't do that today because solar panels are heavy,
6
19000
3000
00:46
expensive and very inefficient.
7
22000
2000
00:48
There are nano-engineered designs,
8
24000
3000
00:51
which at least have been analyzed theoretically,
9
27000
2000
00:53
that show the potential to be very lightweight,
10
29000
2000
00:55
very inexpensive, very efficient,
11
31000
2000
00:57
and we'd be able to actually provide all of our energy needs in this renewable way.
12
33000
4000
01:01
Nano-engineered fuel cells
13
37000
2000
01:03
could provide the energy where it's needed.
14
39000
3000
01:06
That's a key trend, which is decentralization,
15
42000
2000
01:08
moving from centralized nuclear power plants and
16
44000
3000
01:11
liquid natural gas tankers
17
47000
2000
01:13
to decentralized resources that are environmentally more friendly,
18
49000
4000
01:17
a lot more efficient
19
53000
3000
01:20
and capable and safe from disruption.
20
56000
4000
01:24
Bono spoke very eloquently,
21
60000
2000
01:26
that we have the tools, for the first time,
22
62000
4000
01:30
to address age-old problems of disease and poverty.
23
66000
4000
01:34
Most regions of the world are moving in that direction.
24
70000
4000
01:38
In 1990, in East Asia and the Pacific region,
25
74000
4000
01:42
there were 500 million people living in poverty --
26
78000
2000
01:44
that number now is under 200 million.
27
80000
3000
01:47
The World Bank projects by 2011, it will be under 20 million,
28
83000
3000
01:50
which is a reduction of 95 percent.
29
86000
3000
01:53
I did enjoy Bono's comment
30
89000
3000
01:56
linking Haight-Ashbury to Silicon Valley.
31
92000
4000
02:00
Being from the Massachusetts high-tech community myself,
32
96000
3000
02:03
I'd point out that we were hippies also in the 1960s,
33
99000
4000
02:08
although we hung around Harvard Square.
34
104000
3000
02:11
But we do have the potential to overcome disease and poverty,
35
107000
5000
02:16
and I'm going to talk about those issues, if we have the will.
36
112000
3000
02:19
Kevin Kelly talked about the acceleration of technology.
37
115000
3000
02:22
That's been a strong interest of mine,
38
118000
3000
02:25
and a theme that I've developed for some 30 years.
39
121000
3000
02:28
I realized that my technologies had to make sense when I finished a project.
40
124000
5000
02:33
That invariably, the world was a different place
41
129000
3000
02:36
when I would introduce a technology.
42
132000
2000
02:38
And, I noticed that most inventions fail,
43
134000
2000
02:40
not because the R&D department can't get it to work --
44
136000
3000
02:43
if you look at most business plans, they will actually succeed
45
139000
3000
02:46
if given the opportunity to build what they say they're going to build --
46
142000
4000
02:50
and 90 percent of those projects or more will fail, because the timing is wrong --
47
146000
3000
02:53
not all the enabling factors will be in place when they're needed.
48
149000
3000
02:56
So I began to be an ardent student of technology trends,
49
152000
4000
03:00
and track where technology would be at different points in time,
50
156000
3000
03:03
and began to build the mathematical models of that.
51
159000
3000
03:06
It's kind of taken on a life of its own.
52
162000
2000
03:08
I've got a group of 10 people that work with me to gather data
53
164000
3000
03:11
on key measures of technology in many different areas, and we build models.
54
167000
5000
03:16
And you'll hear people say, well, we can't predict the future.
55
172000
3000
03:19
And if you ask me,
56
175000
2000
03:21
will the price of Google be higher or lower than it is today three years from now,
57
177000
3000
03:24
that's very hard to say.
58
180000
2000
03:26
Will WiMax CDMA G3
59
182000
3000
03:29
be the wireless standard three years from now? That's hard to say.
60
185000
2000
03:31
But if you ask me, what will it cost
61
187000
2000
03:33
for one MIPS of computing in 2010,
62
189000
3000
03:36
or the cost to sequence a base pair of DNA in 2012,
63
192000
3000
03:39
or the cost of sending a megabyte of data wirelessly in 2014,
64
195000
4000
03:43
it turns out that those are very predictable.
65
199000
3000
03:46
There are remarkably smooth exponential curves
66
202000
2000
03:48
that govern price performance, capacity, bandwidth.
67
204000
3000
03:51
And I'm going to show you a small sample of this,
68
207000
2000
03:53
but there's really a theoretical reason
69
209000
2000
03:55
why technology develops in an exponential fashion.
70
211000
5000
04:00
And a lot of people, when they think about the future, think about it linearly.
71
216000
2000
04:02
They think they're going to continue
72
218000
2000
04:04
to develop a problem
73
220000
2000
04:06
or address a problem using today's tools,
74
222000
3000
04:09
at today's pace of progress,
75
225000
2000
04:11
and fail to take into consideration this exponential growth.
76
227000
4000
04:15
The Genome Project was a controversial project in 1990.
77
231000
3000
04:18
We had our best Ph.D. students,
78
234000
2000
04:20
our most advanced equipment around the world,
79
236000
2000
04:22
we got 1/10,000th of the project done,
80
238000
2000
04:24
so how're we going to get this done in 15 years?
81
240000
2000
04:26
And 10 years into the project,
82
242000
3000
04:30
the skeptics were still going strong -- says, "You're two-thirds through this project,
83
246000
2000
04:32
and you've managed to only sequence
84
248000
2000
04:34
a very tiny percentage of the whole genome."
85
250000
3000
04:37
But it's the nature of exponential growth
86
253000
2000
04:39
that once it reaches the knee of the curve, it explodes.
87
255000
2000
04:41
Most of the project was done in the last
88
257000
2000
04:43
few years of the project.
89
259000
2000
04:45
It took us 15 years to sequence HIV --
90
261000
2000
04:47
we sequenced SARS in 31 days.
91
263000
2000
04:49
So we are gaining the potential to overcome these problems.
92
265000
4000
04:53
I'm going to show you just a few examples
93
269000
2000
04:55
of how pervasive this phenomena is.
94
271000
3000
04:58
The actual paradigm-shift rate, the rate of adopting new ideas,
95
274000
4000
05:02
is doubling every decade, according to our models.
96
278000
3000
05:05
These are all logarithmic graphs,
97
281000
3000
05:08
so as you go up the levels it represents, generally multiplying by factor of 10 or 100.
98
284000
3000
05:11
It took us half a century to adopt the telephone,
99
287000
3000
05:14
the first virtual-reality technology.
100
290000
3000
05:17
Cell phones were adopted in about eight years.
101
293000
2000
05:19
If you put different communication technologies
102
295000
3000
05:22
on this logarithmic graph,
103
298000
2000
05:24
television, radio, telephone
104
300000
2000
05:26
were adopted in decades.
105
302000
2000
05:28
Recent technologies -- like the PC, the web, cell phones --
106
304000
3000
05:31
were under a decade.
107
307000
2000
05:33
Now this is an interesting chart,
108
309000
2000
05:35
and this really gets at the fundamental reason why
109
311000
2000
05:37
an evolutionary process -- and both biology and technology are evolutionary processes --
110
313000
4000
05:41
accelerate.
111
317000
2000
05:43
They work through interaction -- they create a capability,
112
319000
3000
05:46
and then it uses that capability to bring on the next stage.
113
322000
3000
05:49
So the first step in biological evolution,
114
325000
3000
05:52
the evolution of DNA -- actually it was RNA came first --
115
328000
2000
05:54
took billions of years,
116
330000
2000
05:56
but then evolution used that information-processing backbone
117
332000
3000
05:59
to bring on the next stage.
118
335000
2000
06:01
So the Cambrian Explosion, when all the body plans of the animals were evolved,
119
337000
3000
06:04
took only 10 million years. It was 200 times faster.
120
340000
4000
06:08
And then evolution used those body plans
121
344000
2000
06:10
to evolve higher cognitive functions,
122
346000
2000
06:12
and biological evolution kept accelerating.
123
348000
2000
06:14
It's an inherent nature of an evolutionary process.
124
350000
3000
06:17
So Homo sapiens, the first technology-creating species,
125
353000
3000
06:20
the species that combined a cognitive function
126
356000
2000
06:22
with an opposable appendage --
127
358000
2000
06:24
and by the way, chimpanzees don't really have a very good opposable thumb --
128
360000
4000
06:28
so we could actually manipulate our environment with a power grip
129
364000
2000
06:30
and fine motor coordination,
130
366000
2000
06:32
and use our mental models to actually change the world
131
368000
2000
06:34
and bring on technology.
132
370000
2000
06:36
But anyway, the evolution of our species took hundreds of thousands of years,
133
372000
3000
06:39
and then working through interaction,
134
375000
2000
06:41
evolution used, essentially,
135
377000
2000
06:43
the technology-creating species to bring on the next stage,
136
379000
3000
06:46
which were the first steps in technological evolution.
137
382000
3000
06:49
And the first step took tens of thousands of years --
138
385000
3000
06:52
stone tools, fire, the wheel -- kept accelerating.
139
388000
3000
06:55
We always used then the latest generation of technology
140
391000
2000
06:57
to create the next generation.
141
393000
2000
06:59
Printing press took a century to be adopted;
142
395000
2000
07:01
the first computers were designed pen-on-paper -- now we use computers.
143
397000
4000
07:05
And we've had a continual acceleration of this process.
144
401000
3000
07:08
Now by the way, if you look at this on a linear graph, it looks like everything has just happened,
145
404000
3000
07:11
but some observer says, "Well, Kurzweil just put points on this graph
146
407000
6000
07:17
that fall on that straight line."
147
413000
2000
07:19
So, I took 15 different lists from key thinkers,
148
415000
3000
07:22
like the Encyclopedia Britannica, the Museum of Natural History, Carl Sagan's Cosmic Calendar
149
418000
4000
07:26
on the same -- and these people were not trying to make my point;
150
422000
3000
07:29
these were just lists in reference works,
151
425000
2000
07:31
and I think that's what they thought the key events were
152
427000
3000
07:34
in biological evolution and technological evolution.
153
430000
3000
07:37
And again, it forms the same straight line. You have a little bit of thickening in the line
154
433000
3000
07:40
because people do have disagreements, what the key points are,
155
436000
3000
07:43
there's differences of opinion when agriculture started,
156
439000
2000
07:45
or how long the Cambrian Explosion took.
157
441000
3000
07:48
But you see a very clear trend.
158
444000
2000
07:50
There's a basic, profound acceleration of this evolutionary process.
159
446000
5000
07:55
Information technologies double their capacity, price performance, bandwidth,
160
451000
5000
08:00
every year.
161
456000
2000
08:02
And that's a very profound explosion of exponential growth.
162
458000
4000
08:06
A personal experience, when I was at MIT --
163
462000
2000
08:08
computer taking up about the size of this room,
164
464000
2000
08:10
less powerful than the computer in your cell phone.
165
466000
5000
08:15
But Moore's Law, which is very often identified with this exponential growth,
166
471000
4000
08:19
is just one example of many, because it's basically
167
475000
2000
08:21
a property of the evolutionary process of technology.
168
477000
5000
08:26
I put 49 famous computers on this logarithmic graph --
169
482000
3000
08:29
by the way, a straight line on a logarithmic graph is exponential growth --
170
485000
4000
08:33
that's another exponential.
171
489000
2000
08:35
It took us three years to double our price performance of computing in 1900,
172
491000
3000
08:38
two years in the middle; we're now doubling it every one year.
173
494000
3000
08:42
And that's exponential growth through five different paradigms.
174
498000
3000
08:45
Moore's Law was just the last part of that,
175
501000
2000
08:47
where we were shrinking transistors on an integrated circuit,
176
503000
3000
08:50
but we had electro-mechanical calculators,
177
506000
3000
08:53
relay-based computers that cracked the German Enigma Code,
178
509000
2000
08:55
vacuum tubes in the 1950s predicted the election of Eisenhower,
179
511000
4000
08:59
discreet transistors used in the first space flights
180
515000
3000
09:02
and then Moore's Law.
181
518000
2000
09:04
Every time one paradigm ran out of steam,
182
520000
2000
09:06
another paradigm came out of left field to continue the exponential growth.
183
522000
3000
09:09
They were shrinking vacuum tubes, making them smaller and smaller.
184
525000
3000
09:12
That hit a wall. They couldn't shrink them and keep the vacuum.
185
528000
3000
09:15
Whole different paradigm -- transistors came out of the woodwork.
186
531000
2000
09:17
In fact, when we see the end of the line for a particular paradigm,
187
533000
3000
09:20
it creates research pressure to create the next paradigm.
188
536000
4000
09:24
And because we've been predicting the end of Moore's Law
189
540000
3000
09:27
for quite a long time -- the first prediction said 2002, until now it says 2022.
190
543000
3000
09:30
But by the teen years,
191
546000
3000
09:33
the features of transistors will be a few atoms in width,
192
549000
3000
09:36
and we won't be able to shrink them any more.
193
552000
2000
09:38
That'll be the end of Moore's Law, but it won't be the end of
194
554000
3000
09:41
the exponential growth of computing, because chips are flat.
195
557000
2000
09:43
We live in a three-dimensional world; we might as well use the third dimension.
196
559000
3000
09:46
We will go into the third dimension
197
562000
2000
09:48
and there's been tremendous progress, just in the last few years,
198
564000
3000
09:51
of getting three-dimensional, self-organizing molecular circuits to work.
199
567000
4000
09:55
We'll have those ready well before Moore's Law runs out of steam.
200
571000
7000
10:02
Supercomputers -- same thing.
201
578000
2000
10:05
Processor performance on Intel chips,
202
581000
3000
10:08
the average price of a transistor --
203
584000
3000
10:11
1968, you could buy one transistor for a dollar.
204
587000
3000
10:14
You could buy 10 million in 2002.
205
590000
3000
10:17
It's pretty remarkable how smooth
206
593000
3000
10:20
an exponential process that is.
207
596000
2000
10:22
I mean, you'd think this is the result of some tabletop experiment,
208
598000
3000
10:26
but this is the result of worldwide chaotic behavior --
209
602000
3000
10:29
countries accusing each other of dumping products,
210
605000
2000
10:31
IPOs, bankruptcies, marketing programs.
211
607000
2000
10:33
You would think it would be a very erratic process,
212
609000
3000
10:36
and you have a very smooth
213
612000
2000
10:38
outcome of this chaotic process.
214
614000
2000
10:40
Just as we can't predict
215
616000
2000
10:42
what one molecule in a gas will do --
216
618000
2000
10:44
it's hopeless to predict a single molecule --
217
620000
3000
10:47
yet we can predict the properties of the whole gas,
218
623000
2000
10:49
using thermodynamics, very accurately.
219
625000
3000
10:52
It's the same thing here. We can't predict any particular project,
220
628000
3000
10:55
but the result of this whole worldwide,
221
631000
2000
10:57
chaotic, unpredictable activity of competition
222
633000
5000
11:02
and the evolutionary process of technology is very predictable.
223
638000
3000
11:05
And we can predict these trends far into the future.
224
641000
3000
11:10
Unlike Gertrude Stein's roses,
225
646000
2000
11:12
it's not the case that a transistor is a transistor.
226
648000
2000
11:14
As we make them smaller and less expensive,
227
650000
2000
11:16
the electrons have less distance to travel.
228
652000
2000
11:18
They're faster, so you've got exponential growth in the speed of transistors,
229
654000
4000
11:22
so the cost of a cycle of one transistor
230
658000
4000
11:26
has been coming down with a halving rate of 1.1 years.
231
662000
3000
11:29
You add other forms of innovation and processor design,
232
665000
3000
11:32
you get a doubling of price performance of computing every one year.
233
668000
4000
11:36
And that's basically deflation --
234
672000
3000
11:39
50 percent deflation.
235
675000
2000
11:41
And it's not just computers. I mean, it's true of DNA sequencing;
236
677000
3000
11:44
it's true of brain scanning;
237
680000
2000
11:46
it's true of the World Wide Web. I mean, anything that we can quantify,
238
682000
2000
11:48
we have hundreds of different measurements
239
684000
3000
11:51
of different, information-related measurements --
240
687000
3000
11:54
capacity, adoption rates --
241
690000
2000
11:56
and they basically double every 12, 13, 15 months,
242
692000
3000
11:59
depending on what you're looking at.
243
695000
2000
12:01
In terms of price performance, that's a 40 to 50 percent deflation rate.
244
697000
4000
12:06
And economists have actually started worrying about that.
245
702000
2000
12:08
We had deflation during the Depression,
246
704000
2000
12:10
but that was collapse of the money supply,
247
706000
2000
12:12
collapse of consumer confidence, a completely different phenomena.
248
708000
3000
12:15
This is due to greater productivity,
249
711000
2000
12:18
but the economist says, "But there's no way you're going to be able to keep up with that.
250
714000
2000
12:20
If you have 50 percent deflation, people may increase their volume
251
716000
3000
12:23
30, 40 percent, but they won't keep up with it."
252
719000
2000
12:25
But what we're actually seeing is that
253
721000
2000
12:27
we actually more than keep up with it.
254
723000
2000
12:29
We've had 28 percent per year compounded growth in dollars
255
725000
3000
12:32
in information technology over the last 50 years.
256
728000
3000
12:35
I mean, people didn't build iPods for 10,000 dollars 10 years ago.
257
731000
4000
12:39
As the price performance makes new applications feasible,
258
735000
3000
12:42
new applications come to the market.
259
738000
2000
12:44
And this is a very widespread phenomena.
260
740000
3000
12:47
Magnetic data storage --
261
743000
2000
12:49
that's not Moore's Law, it's shrinking magnetic spots,
262
745000
3000
12:52
different engineers, different companies, same exponential process.
263
748000
4000
12:56
A key revolution is that we're understanding our own biology
264
752000
4000
13:00
in these information terms.
265
756000
2000
13:02
We're understanding the software programs
266
758000
2000
13:04
that make our body run.
267
760000
2000
13:06
These were evolved in very different times --
268
762000
2000
13:08
we'd like to actually change those programs.
269
764000
2000
13:10
One little software program, called the fat insulin receptor gene,
270
766000
2000
13:12
basically says, "Hold onto every calorie,
271
768000
2000
13:14
because the next hunting season may not work out so well."
272
770000
4000
13:18
That was in the interests of the species tens of thousands of years ago.
273
774000
3000
13:21
We'd like to actually turn that program off.
274
777000
3000
13:24
They tried that in animals, and these mice ate ravenously
275
780000
3000
13:27
and remained slim and got the health benefits of being slim.
276
783000
2000
13:29
They didn't get diabetes; they didn't get heart disease;
277
785000
3000
13:32
they lived 20 percent longer; they got the health benefits of caloric restriction
278
788000
3000
13:35
without the restriction.
279
791000
2000
13:37
Four or five pharmaceutical companies have noticed this,
280
793000
3000
13:40
felt that would be
281
796000
3000
13:43
interesting drug for the human market,
282
799000
3000
13:46
and that's just one of the 30,000 genes
283
802000
2000
13:48
that affect our biochemistry.
284
804000
3000
13:51
We were evolved in an era where it wasn't in the interests of people
285
807000
3000
13:54
at the age of most people at this conference, like myself,
286
810000
3000
13:57
to live much longer, because we were using up the precious resources
287
813000
4000
14:01
which were better deployed towards the children
288
817000
1000
14:02
and those caring for them.
289
818000
2000
14:04
So, life -- long lifespans --
290
820000
2000
14:06
like, that is to say, much more than 30 --
291
822000
2000
14:08
weren't selected for,
292
824000
3000
14:11
but we are learning to actually manipulate
293
827000
3000
14:14
and change these software programs
294
830000
2000
14:16
through the biotechnology revolution.
295
832000
2000
14:18
For example, we can inhibit genes now with RNA interference.
296
834000
4000
14:22
There are exciting new forms of gene therapy
297
838000
2000
14:24
that overcome the problem of placing the genetic material
298
840000
2000
14:26
in the right place on the chromosome.
299
842000
2000
14:28
There's actually a -- for the first time now,
300
844000
3000
14:31
something going to human trials, that actually cures pulmonary hypertension --
301
847000
3000
14:34
a fatal disease -- using gene therapy.
302
850000
3000
14:37
So we'll have not just designer babies, but designer baby boomers.
303
853000
3000
14:40
And this technology is also accelerating.
304
856000
3000
14:43
It cost 10 dollars per base pair in 1990,
305
859000
3000
14:46
then a penny in 2000.
306
862000
2000
14:48
It's now under a 10th of a cent.
307
864000
2000
14:50
The amount of genetic data --
308
866000
2000
14:52
basically this shows that smooth exponential growth
309
868000
3000
14:55
doubled every year,
310
871000
2000
14:57
enabling the genome project to be completed.
311
873000
3000
15:00
Another major revolution: the communications revolution.
312
876000
3000
15:03
The price performance, bandwidth, capacity of communications measured many different ways;
313
879000
5000
15:08
wired, wireless is growing exponentially.
314
884000
3000
15:11
The Internet has been doubling in power and continues to,
315
887000
3000
15:14
measured many different ways.
316
890000
2000
15:16
This is based on the number of hosts.
317
892000
2000
15:18
Miniaturization -- we're shrinking the size of technology
318
894000
2000
15:20
at an exponential rate,
319
896000
2000
15:22
both wired and wireless.
320
898000
2000
15:24
These are some designs from Eric Drexler's book --
321
900000
4000
15:28
which we're now showing are feasible
322
904000
2000
15:30
with super-computing simulations,
323
906000
2000
15:32
where actually there are scientists building
324
908000
2000
15:34
molecule-scale robots.
325
910000
2000
15:36
One has one that actually walks with a surprisingly human-like gait,
326
912000
2000
15:38
that's built out of molecules.
327
914000
3000
15:41
There are little machines doing things in experimental bases.
328
917000
4000
15:45
The most exciting opportunity
329
921000
3000
15:48
is actually to go inside the human body
330
924000
2000
15:50
and perform therapeutic and diagnostic functions.
331
926000
3000
15:53
And this is less futuristic than it may sound.
332
929000
2000
15:55
These things have already been done in animals.
333
931000
2000
15:57
There's one nano-engineered device that cures type 1 diabetes. It's blood cell-sized.
334
933000
4000
16:01
They put tens of thousands of these
335
937000
2000
16:03
in the blood cell -- they tried this in rats --
336
939000
2000
16:05
it lets insulin out in a controlled fashion,
337
941000
2000
16:07
and actually cures type 1 diabetes.
338
943000
2000
16:09
What you're watching is a design
339
945000
3000
16:12
of a robotic red blood cell,
340
948000
2000
16:14
and it does bring up the issue that our biology
341
950000
2000
16:16
is actually very sub-optimal,
342
952000
2000
16:18
even though it's remarkable in its intricacy.
343
954000
3000
16:21
Once we understand its principles of operation,
344
957000
3000
16:24
and the pace with which we are reverse-engineering biology is accelerating,
345
960000
3000
16:28
we can actually design these things to be
346
964000
2000
16:30
thousands of times more capable.
347
966000
2000
16:32
An analysis of this respirocyte, designed by Rob Freitas,
348
968000
4000
16:37
indicates if you replace 10 percent of your red blood cells with these robotic versions,
349
973000
2000
16:40
you could do an Olympic sprint for 15 minutes without taking a breath.
350
976000
3000
16:43
You could sit at the bottom of your pool for four hours --
351
979000
3000
16:46
so, "Honey, I'm in the pool," will take on a whole new meaning.
352
982000
4000
16:50
It will be interesting to see what we do in our Olympic trials.
353
986000
2000
16:52
Presumably we'll ban them,
354
988000
2000
16:54
but then we'll have the specter of teenagers in their high schools gyms
355
990000
2000
16:56
routinely out-performing the Olympic athletes.
356
992000
3000
17:01
Freitas has a design for a robotic white blood cell.
357
997000
3000
17:04
These are 2020-circa scenarios,
358
1000000
4000
17:08
but they're not as futuristic as it may sound.
359
1004000
2000
17:10
There are four major conferences on building blood cell-sized devices;
360
1006000
4000
17:14
there are many experiments in animals.
361
1010000
2000
17:16
There's actually one going into human trial,
362
1012000
2000
17:18
so this is feasible technology.
363
1014000
3000
17:22
If we come back to our exponential growth of computing,
364
1018000
2000
17:24
1,000 dollars of computing is now somewhere between an insect and a mouse brain.
365
1020000
3000
17:27
It will intersect human intelligence
366
1023000
3000
17:30
in terms of capacity in the 2020s,
367
1026000
3000
17:33
but that'll be the hardware side of the equation.
368
1029000
2000
17:35
Where will we get the software?
369
1031000
2000
17:37
Well, it turns out we can see inside the human brain,
370
1033000
2000
17:39
and in fact not surprisingly,
371
1035000
2000
17:41
the spatial and temporal resolution of brain scanning is doubling every year.
372
1037000
4000
17:45
And with the new generation of scanning tools,
373
1041000
2000
17:47
for the first time we can actually see
374
1043000
2000
17:49
individual inter-neural fibers
375
1045000
2000
17:51
and see them processing and signaling in real time --
376
1047000
3000
17:54
but then the question is, OK, we can get this data now,
377
1050000
2000
17:56
but can we understand it?
378
1052000
2000
17:58
Doug Hofstadter wonders, well, maybe our intelligence
379
1054000
3000
18:01
just isn't great enough to understand our intelligence,
380
1057000
3000
18:04
and if we were smarter, well, then our brains would be that much more complicated,
381
1060000
3000
18:07
and we'd never catch up to it.
382
1063000
2000
18:10
It turns out that we can understand it.
383
1066000
3000
18:13
This is a block diagram of
384
1069000
3000
18:16
a model and simulation of the human auditory cortex
385
1072000
4000
18:20
that actually works quite well --
386
1076000
2000
18:22
in applying psychoacoustic tests, gets very similar results to human auditory perception.
387
1078000
2000
18:26
There's another simulation of the cerebellum --
388
1082000
3000
18:29
that's more than half the neurons in the brain --
389
1085000
2000
18:31
again, works very similarly to human skill formation.
390
1087000
3000
18:35
This is at an early stage, but you can show
391
1091000
3000
18:38
with the exponential growth of the amount of information about the brain
392
1094000
3000
18:41
and the exponential improvement
393
1097000
2000
18:43
in the resolution of brain scanning,
394
1099000
2000
18:45
we will succeed in reverse-engineering the human brain
395
1101000
3000
18:48
by the 2020s.
396
1104000
2000
18:50
We've already had very good models and simulation of about 15 regions
397
1106000
3000
18:53
out of the several hundred.
398
1109000
3000
18:56
All of this is driving
399
1112000
2000
18:58
exponentially growing economic progress.
400
1114000
2000
19:00
We've had productivity go from 30 dollars to 150 dollars per hour
401
1116000
3000
19:05
of labor in the last 50 years.
402
1121000
2000
19:07
E-commerce has been growing exponentially. It's now a trillion dollars.
403
1123000
3000
19:10
You might wonder, well, wasn't there a boom and a bust?
404
1126000
2000
19:12
That was strictly a capital-markets phenomena.
405
1128000
2000
19:14
Wall Street noticed that this was a revolutionary technology, which it was,
406
1130000
4000
19:18
but then six months later, when it hadn't revolutionized all business models,
407
1134000
3000
19:21
they figured, well, that was wrong,
408
1137000
2000
19:23
and then we had this bust.
409
1139000
2000
19:26
All right, this is a technology
410
1142000
2000
19:28
that we put together using some of the technologies we're involved in.
411
1144000
3000
19:31
This will be a routine feature in a cell phone.
412
1147000
4000
19:35
It would be able to translate from one language to another.
413
1151000
2000
19:47
So let me just end with a couple of scenarios.
414
1163000
2000
19:49
By 2010 computers will disappear.
415
1165000
3000
19:53
They'll be so small, they'll be embedded in our clothing, in our environment.
416
1169000
3000
19:56
Images will be written directly to our retina,
417
1172000
2000
19:58
providing full-immersion virtual reality,
418
1174000
2000
20:00
augmented real reality. We'll be interacting with virtual personalities.
419
1176000
3000
20:04
But if we go to 2029, we really have the full maturity of these trends,
420
1180000
4000
20:08
and you have to appreciate how many turns of the screw
421
1184000
3000
20:11
in terms of generations of technology, which are getting faster and faster, we'll have at that point.
422
1187000
4000
20:15
I mean, we will have two-to-the-25th-power
423
1191000
2000
20:17
greater price performance, capacity and bandwidth
424
1193000
3000
20:20
of these technologies, which is pretty phenomenal.
425
1196000
2000
20:22
It'll be millions of times more powerful than it is today.
426
1198000
2000
20:24
We'll have completed the reverse-engineering of the human brain,
427
1200000
2000
20:27
1,000 dollars of computing will be far more powerful
428
1203000
3000
20:30
than the human brain in terms of basic raw capacity.
429
1206000
4000
20:34
Computers will combine
430
1210000
2000
20:36
the subtle pan-recognition powers
431
1212000
2000
20:38
of human intelligence with ways in which machines are already superior,
432
1214000
3000
20:41
in terms of doing analytic thinking,
433
1217000
2000
20:43
remembering billions of facts accurately.
434
1219000
2000
20:45
Machines can share their knowledge very quickly.
435
1221000
2000
20:47
But it's not just an alien invasion of intelligent machines.
436
1223000
5000
20:52
We are going to merge with our technology.
437
1228000
2000
20:54
These nano-bots I mentioned
438
1230000
2000
20:56
will first be used for medical and health applications:
439
1232000
4000
21:00
cleaning up the environment, providing powerful fuel cells
440
1236000
3000
21:03
and widely distributed decentralized solar panels and so on in the environment.
441
1239000
5000
21:08
But they'll also go inside our brain,
442
1244000
2000
21:10
interact with our biological neurons.
443
1246000
2000
21:12
We've demonstrated the key principles of being able to do this.
444
1248000
3000
21:15
So, for example,
445
1251000
2000
21:17
full-immersion virtual reality from within the nervous system,
446
1253000
2000
21:19
the nano-bots shut down the signals coming from your real senses,
447
1255000
3000
21:22
replace them with the signals that your brain would be receiving
448
1258000
3000
21:25
if you were in the virtual environment,
449
1261000
2000
21:27
and then it'll feel like you're in that virtual environment.
450
1263000
2000
21:29
You can go there with other people, have any kind of experience
451
1265000
2000
21:31
with anyone involving all of the senses.
452
1267000
2000
21:34
"Experience beamers," I call them, will put their whole flow of sensory experiences
453
1270000
3000
21:37
in the neurological correlates of their emotions out on the Internet.
454
1273000
3000
21:40
You can plug in and experience what it's like to be someone else.
455
1276000
3000
21:43
But most importantly,
456
1279000
2000
21:45
it'll be a tremendous expansion
457
1281000
2000
21:47
of human intelligence through this direct merger with our technology,
458
1283000
4000
21:51
which in some sense we're doing already.
459
1287000
2000
21:53
We routinely do intellectual feats
460
1289000
2000
21:55
that would be impossible without our technology.
461
1291000
2000
21:57
Human life expectancy is expanding. It was 37 in 1800,
462
1293000
3000
22:00
and with this sort of biotechnology, nano-technology revolutions,
463
1296000
5000
22:05
this will move up very rapidly
464
1301000
2000
22:07
in the years ahead.
465
1303000
2000
22:09
My main message is that progress in technology
466
1305000
4000
22:13
is exponential, not linear.
467
1309000
3000
22:16
Many -- even scientists -- assume a linear model,
468
1312000
4000
22:20
so they'll say, "Oh, it'll be hundreds of years
469
1316000
2000
22:22
before we have self-replicating nano-technology assembly
470
1318000
3000
22:25
or artificial intelligence."
471
1321000
2000
22:27
If you really look at the power of exponential growth,
472
1323000
3000
22:30
you'll see that these things are pretty soon at hand.
473
1326000
3000
22:33
And information technology is increasingly encompassing
474
1329000
3000
22:36
all of our lives, from our music to our manufacturing
475
1332000
4000
22:40
to our biology to our energy to materials.
476
1336000
4000
22:44
We'll be able to manufacture almost anything we need in the 2020s,
477
1340000
3000
22:47
from information, in very inexpensive raw materials,
478
1343000
2000
22:49
using nano-technology.
479
1345000
3000
22:52
These are very powerful technologies.
480
1348000
2000
22:54
They both empower our promise and our peril.
481
1350000
4000
22:58
So we have to have the will to apply them to the right problems.
482
1354000
3000
23:01
Thank you very much.
483
1357000
1000
23:02
(Applause)
484
1358000
1000
ABOUT THE SPEAKER
Ray Kurzweil - Inventor, futuristRay Kurzweil is an engineer who has radically advanced the fields of speech, text and audio technology. He's revered for his dizzying -- yet convincing -- writing on the advance of technology, the limits of biology and the future of the human species.
Why you should listen
Inventor, entrepreneur, visionary, Ray Kurzweil's accomplishments read as a startling series of firsts -- a litany of technological breakthroughs we've come to take for granted. Kurzweil invented the first optical character recognition (OCR) software for transforming the written word into data, the first print-to-speech software for the blind, the first text-to-speech synthesizer, and the first music synthesizer capable of recreating the grand piano and other orchestral instruments, and the first commercially marketed large-vocabulary speech recognition.
Yet his impact as a futurist and philosopher is no less significant. In his best-selling books, which include How to Create a Mind, The Age of Spiritual Machines, The Singularity Is Near: When Humans Transcend Biology, Kurzweil depicts in detail a portrait of the human condition over the next few decades, as accelerating technologies forever blur the line between human and machine.
In 2009, he unveiled Singularity University, an institution that aims to "assemble, educate and inspire leaders who strive to understand and facilitate the development of exponentially advancing technologies." He is a Director of Engineering at Google, where he heads up a team developing machine intelligence and natural language comprehension.
Ray Kurzweil | Speaker | TED.com