1
00:00:01,120 --> 00:00:03,490
The following content is
provided under a Creative

2
00:00:03,490 --> 00:00:04,880
Commons license.

3
00:00:04,880 --> 00:00:07,090
Your support will help
MIT OpenCourseWare

4
00:00:07,090 --> 00:00:11,180
continue to offer high-quality
educational resources for free.

5
00:00:11,180 --> 00:00:13,750
To make a donation or to
view additional materials

6
00:00:13,750 --> 00:00:17,680
from hundreds of MIT courses,
visit MIT OpenCourseWare

7
00:00:17,680 --> 00:00:18,586
at OCW.MIT.edu.

8
00:00:23,752 --> 00:00:26,210
PROFESSOR: I'm excited, because
I'm going to get the chance

9
00:00:26,210 --> 00:00:30,260
to co-teach today with--

10
00:00:30,260 --> 00:00:32,479
I would say one of the
students, but it's not really

11
00:00:32,479 --> 00:00:33,312
one of the students.

12
00:00:33,312 --> 00:00:37,370
But Larry Lessig has consented
to join us in a few minutes.

13
00:00:37,370 --> 00:00:39,800
So I'm going to be
breezing through

14
00:00:39,800 --> 00:00:42,390
a little faster than usual.

15
00:00:42,390 --> 00:00:46,320
And then we'll co-teach this.

16
00:00:46,320 --> 00:00:47,360
Blockchain and Money.

17
00:00:47,360 --> 00:00:47,960
Here we are.

18
00:00:47,960 --> 00:00:49,520
We're at smart contracts.

19
00:00:49,520 --> 00:00:51,320
Everybody seems
to be coming back,

20
00:00:51,320 --> 00:00:55,670
which is a sign of your interest
more than it is in my teaching.

21
00:00:55,670 --> 00:00:57,990
But I thank you for being here.

22
00:00:57,990 --> 00:00:59,750
We made it through
the last three classes

23
00:00:59,750 --> 00:01:03,790
together on bitcoin and the
basics of that technology.

24
00:01:03,790 --> 00:01:05,630
And in one class,
we're going to try

25
00:01:05,630 --> 00:01:08,300
to chew off a bit
on smart contracts,

26
00:01:08,300 --> 00:01:13,370
both the technology side, a
little bit on the markets,

27
00:01:13,370 --> 00:01:18,260
and then the law that Larry
is going to take us through.

28
00:01:18,260 --> 00:01:21,453
And so as I said,
I want to start

29
00:01:21,453 --> 00:01:23,120
with a little
administrative, because we

30
00:01:23,120 --> 00:01:24,230
are on the sixth class.

31
00:01:24,230 --> 00:01:28,210
I'm going to review a bit about
the projects very quickly.

32
00:01:28,210 --> 00:01:31,730
We're going to do smart
contracts, the design.

33
00:01:31,730 --> 00:01:33,590
What design features?

34
00:01:33,590 --> 00:01:35,450
Yes, we're going to
go back a little bit

35
00:01:35,450 --> 00:01:39,340
about hash functions and
Merkle trees but not too much.

36
00:01:39,340 --> 00:01:43,310
DApps, which are basically
decentralized applications,

37
00:01:43,310 --> 00:01:45,430
and token sales.

38
00:01:45,430 --> 00:01:47,990
Larry's going to take
us through legal issues,

39
00:01:47,990 --> 00:01:50,330
and then we'll sum it all up.

40
00:01:50,330 --> 00:01:54,240
So let me just, real fast on
the administrative side again--

41
00:01:54,240 --> 00:01:56,840
class participation, 30%.

42
00:01:56,840 --> 00:01:59,510
That's why we're
all here together.

43
00:01:59,510 --> 00:02:02,330
That means, hopefully,
reading the assignments

44
00:02:02,330 --> 00:02:03,710
and participating.

45
00:02:03,710 --> 00:02:06,500
About half of you
have participated,

46
00:02:06,500 --> 00:02:09,199
so I guess I'm going to
go with a little bit more

47
00:02:09,199 --> 00:02:10,940
cold calling starting Thursday.

48
00:02:10,940 --> 00:02:13,190
I'm not going to do a bunch
of it today, because Larry

49
00:02:13,190 --> 00:02:14,480
and I are joining.

50
00:02:14,480 --> 00:02:15,860
So you can ease off.

51
00:02:15,860 --> 00:02:19,550
But really be conscious.

52
00:02:19,550 --> 00:02:22,190
If you haven't been
participating, try to get in

53
00:02:22,190 --> 00:02:25,990
and join the conversation
and discussion.

54
00:02:25,990 --> 00:02:28,790
The two individual write-ups,
I think we have seven.

55
00:02:28,790 --> 00:02:32,700
But it's quite possible some
of you have submitted already.

56
00:02:32,700 --> 00:02:36,890
It's just meant to be
critical business reasoning.

57
00:02:36,890 --> 00:02:40,070
We are in a business school.

58
00:02:40,070 --> 00:02:43,720
To the extent you just
summarize some readings,

59
00:02:43,720 --> 00:02:47,220
I'm sure we'll give
you a pass on that.

60
00:02:47,220 --> 00:02:48,720
But that's not what
I'm looking for.

61
00:02:48,720 --> 00:02:50,690
I'm really looking
for critical reasoning

62
00:02:50,690 --> 00:02:53,380
and thinking from a
business perspective

63
00:02:53,380 --> 00:02:55,700
of, why does this matter?

64
00:02:55,700 --> 00:02:57,620
What are its strengths
or its values

65
00:02:57,620 --> 00:03:01,280
in terms of business reasoning
and critical reasoning?

66
00:03:01,280 --> 00:03:07,540
One by the 10th class, one by
the 23rd, always before class.

67
00:03:07,540 --> 00:03:09,800
What it likely means
is we'll be getting

68
00:03:09,800 --> 00:03:12,200
most of these on the
8th, 9th, and 10th class.

69
00:03:12,200 --> 00:03:14,640
I know how that
works and so forth.

70
00:03:14,640 --> 00:03:15,580
And that's OK.

71
00:03:15,580 --> 00:03:16,460
That's OK.

72
00:03:16,460 --> 00:03:19,220
But I'm just
reminding you of that.

73
00:03:19,220 --> 00:03:21,540
In terms of the
group research paper,

74
00:03:21,540 --> 00:03:27,080
again, Sabrina and Thalita
stood up a Google App--

75
00:03:27,080 --> 00:03:28,970
I think it's in Google, but--

76
00:03:28,970 --> 00:03:34,190
where you can go in and team
up in teams of three or four.

77
00:03:34,190 --> 00:03:37,620
It's not required to do this by
the eighth class next Tuesday,

78
00:03:37,620 --> 00:03:40,400
but I'm strongly
suggesting it to figure out

79
00:03:40,400 --> 00:03:44,300
who your teams are and not sort
of wait till the second half

80
00:03:44,300 --> 00:03:45,590
of the semester.

81
00:03:45,590 --> 00:03:47,330
And I'd like to
encourage everybody

82
00:03:47,330 --> 00:03:50,528
to choose an area for
your use case by midway.

83
00:03:50,528 --> 00:03:52,070
Again, you're not
going to get graded

84
00:03:52,070 --> 00:03:55,010
if it takes you another class
or two to figure it out.

85
00:03:55,010 --> 00:03:59,060
But I just think it's much
better if you know your teams,

86
00:03:59,060 --> 00:04:02,400
you know what your
use case will be.

87
00:04:02,400 --> 00:04:05,450
If it's not about finance--

88
00:04:05,450 --> 00:04:10,070
one group has already asked
me if that's all right.

89
00:04:10,070 --> 00:04:13,550
I just want to know more
detailed what it's going to be.

90
00:04:13,550 --> 00:04:17,180
I'm probably going
to say yes, but it

91
00:04:17,180 --> 00:04:18,589
needs a little
bit of preapproval

92
00:04:18,589 --> 00:04:20,180
if it's outside of finance.

93
00:04:20,180 --> 00:04:22,310
So any questions about
the requirements?

94
00:04:22,310 --> 00:04:23,160
I just wanted to--

95
00:04:25,980 --> 00:04:29,470
So the study questions.

96
00:04:29,470 --> 00:04:33,090
Today's smart contracts
and basically how they

97
00:04:33,090 --> 00:04:36,840
compare to regular contracts.

98
00:04:36,840 --> 00:04:42,900
And what are the tokens that
are used within that ecosystem?

99
00:04:42,900 --> 00:04:46,110
And what are the platforms?

100
00:04:46,110 --> 00:04:51,270
In a very similar sense to the
internet, we have the internet.

101
00:04:51,270 --> 00:04:54,750
And then above the internet,
you have, you might say,

102
00:04:54,750 --> 00:04:55,570
applications--

103
00:04:55,570 --> 00:04:59,200
Facebook and other applications.

104
00:04:59,200 --> 00:05:02,310
Well, this, too, has
a series of platforms

105
00:05:02,310 --> 00:05:05,010
and then decentralized
applications

106
00:05:05,010 --> 00:05:08,810
potentially on top of that.

107
00:05:08,810 --> 00:05:12,643
And basically, a quick touch
on decentralized applications.

108
00:05:12,643 --> 00:05:14,060
Later in the
semester, we're going

109
00:05:14,060 --> 00:05:17,300
to take two sessions on
initial coin offerings.

110
00:05:17,300 --> 00:05:21,980
And thus, we're going to talk
a lot about token economics.

111
00:05:21,980 --> 00:05:24,500
So this will be just
the first taste test,

112
00:05:24,500 --> 00:05:28,660
and then we'll come
back to it later.

113
00:05:28,660 --> 00:05:31,770
And then the readings,
hopefully, everybody

114
00:05:31,770 --> 00:05:34,800
read through "Smart Contracts."

115
00:05:34,800 --> 00:05:37,050
I thought the Chamber
of Digital Commerce,

116
00:05:37,050 --> 00:05:38,850
even though this is
almost two years ago,

117
00:05:38,850 --> 00:05:43,260
this paper was a very
helpful sort of flavor

118
00:05:43,260 --> 00:05:47,400
for what folks in
commerce are thinking

119
00:05:47,400 --> 00:05:49,140
and what developers
were thinking.

120
00:05:49,140 --> 00:05:53,920
And I love that Nick Szabo
wrote the introduction to it.

121
00:05:53,920 --> 00:05:56,670
And it's interesting
to see and question,

122
00:05:56,670 --> 00:05:58,890
why is it that it's
two years later,

123
00:05:58,890 --> 00:06:01,890
and many of those use cases
are still being discussed

124
00:06:01,890 --> 00:06:05,940
but haven't fully been adopted?

125
00:06:05,940 --> 00:06:07,480
And then the "State
of the Dapps,"

126
00:06:07,480 --> 00:06:10,712
and then who are the
competitors to Ethereum?

127
00:06:10,712 --> 00:06:11,670
So don't you feel good?

128
00:06:11,670 --> 00:06:15,540
I haven't asked anybody
a cold question yet.

129
00:06:15,540 --> 00:06:18,810
And then the optional
readings, I don't know.

130
00:06:18,810 --> 00:06:20,040
Did anybody actually go back?

131
00:06:20,040 --> 00:06:20,748
They're optional.

132
00:06:20,748 --> 00:06:24,120
Did anybody go back and read
either Szabo's original piece

133
00:06:24,120 --> 00:06:26,350
from 20 years ago
on smart contracts?

134
00:06:26,350 --> 00:06:30,240
Oh, a few of you are kind
of into that rabbit hole

135
00:06:30,240 --> 00:06:31,580
of blockchain and ether.

136
00:06:34,590 --> 00:06:37,740
Brotish what did you think of--
it's 20 years ago he wrote this

137
00:06:37,740 --> 00:06:38,760
thing on--

138
00:06:38,760 --> 00:06:40,130
Nick Szabo wrote this thing.

139
00:06:40,130 --> 00:06:41,850
AUDIENCE: Actually,
I spent more time

140
00:06:41,850 --> 00:06:43,340
on the Ethereum white paper.

141
00:06:43,340 --> 00:06:43,590
PROFESSOR: All right.

142
00:06:43,590 --> 00:06:46,095
So what did you think about
the Ethereum white paper one?

143
00:06:46,095 --> 00:06:47,887
AUDIENCE: Yeah, I think
it was pretty good.

144
00:06:47,887 --> 00:06:50,140
It gave me a very good
overview of the world potential

145
00:06:50,140 --> 00:06:51,203
of the [INAUDIBLE].

146
00:06:51,203 --> 00:06:52,050
PROFESSOR: Yeah.

147
00:06:52,050 --> 00:06:54,930
And what's interesting is
even if you didn't do it

148
00:06:54,930 --> 00:06:57,210
for the class and you find
yourself more and more

149
00:06:57,210 --> 00:07:01,170
interested in this over the next
couple of months or even later,

150
00:07:01,170 --> 00:07:03,540
going back and reading
the Ethereum white paper,

151
00:07:03,540 --> 00:07:06,840
it's not highly technical
in the first, I don't know,

152
00:07:06,840 --> 00:07:08,400
10 or 15 pages.

153
00:07:08,400 --> 00:07:10,110
It really gives a
history of bitcoin.

154
00:07:10,110 --> 00:07:13,350
It talks about
distributed applications

155
00:07:13,350 --> 00:07:21,040
and largely written at the
time by a 19-year-old, as well.

156
00:07:21,040 --> 00:07:25,840
It's a remarkable thing.

157
00:07:25,840 --> 00:07:29,470
And then one of Larry's
colleague, or maybe it's

158
00:07:29,470 --> 00:07:32,320
two of them, but De
Philippi paper, as well,

159
00:07:32,320 --> 00:07:34,520
on the regulatory issues.

160
00:07:34,520 --> 00:07:36,610
So let me talk a little
bit about smart contracts

161
00:07:36,610 --> 00:07:40,960
and laying groundwork before
Larry gives us the law.

162
00:07:40,960 --> 00:07:44,650
A smart contract is
a set of promises

163
00:07:44,650 --> 00:07:47,350
specified in a digital form.

164
00:07:47,350 --> 00:07:48,680
I'm going to say four things.

165
00:07:48,680 --> 00:07:53,640
It's just a set of
promises in a digital form.

166
00:07:53,640 --> 00:07:57,070
So it's not handwritten out.

167
00:07:57,070 --> 00:07:58,330
It includes protocols.

168
00:07:58,330 --> 00:07:59,825
What's a protocol?

169
00:07:59,825 --> 00:08:00,325
Andrew.

170
00:08:05,955 --> 00:08:07,580
AUDIENCE: Standard
operating procedure.

171
00:08:07,580 --> 00:08:09,770
PROFESSOR: Standard
operating procedure.

172
00:08:09,770 --> 00:08:10,530
I like that.

173
00:08:10,530 --> 00:08:14,450
Anybody else give me
another word for it, maybe?

174
00:08:14,450 --> 00:08:17,610
An algorithm.

175
00:08:17,610 --> 00:08:21,480
So a set of promises
in digital form.

176
00:08:21,480 --> 00:08:25,970
But it can include math,
basically, if you wish

177
00:08:25,970 --> 00:08:30,560
or standard operating
procedures-- if-then statements

178
00:08:30,560 --> 00:08:32,380
and so forth.

179
00:08:32,380 --> 00:08:36,375
And the parties then perform
against these promises.

180
00:08:36,375 --> 00:08:37,000
And guess what?

181
00:08:37,000 --> 00:08:39,640
Nick Szabo wrote that in 1996.

182
00:08:39,640 --> 00:08:41,770
And I thought it was
still probably the best

183
00:08:41,770 --> 00:08:44,140
definition of smart contract.

184
00:08:44,140 --> 00:08:47,440
He coined the
phrase 22 years ago.

185
00:08:47,440 --> 00:08:49,900
He might actually
be Satoshi Nakamoto.

186
00:08:49,900 --> 00:08:52,330
Three of the tables
in here, you all

187
00:08:52,330 --> 00:08:53,770
voted that it was Nick Szabo.

188
00:08:56,500 --> 00:08:58,990
So I thought that's kind
of the best definition

189
00:08:58,990 --> 00:09:02,650
if you've got to
just-- kind of a root.

190
00:09:02,650 --> 00:09:06,490
Now, I would also say, however,
that smart contracts may not

191
00:09:06,490 --> 00:09:07,657
be so smart.

192
00:09:07,657 --> 00:09:09,490
A lot of people have
come to be calling them

193
00:09:09,490 --> 00:09:11,950
dumb contracts that
are just algorithms

194
00:09:11,950 --> 00:09:14,170
that perform a function.

195
00:09:14,170 --> 00:09:17,920
So don't think of them as
artificial intelligence.

196
00:09:17,920 --> 00:09:23,710
That's another class provided
next semester by Simon Johnson.

197
00:09:23,710 --> 00:09:28,900
Think of them almost
as just dumb contracts.

198
00:09:28,900 --> 00:09:31,050
And in a sense,
they're mechanizing

199
00:09:31,050 --> 00:09:35,280
what might otherwise be done
amongst and between humans.

200
00:09:35,280 --> 00:09:38,280
And smart contracts may or
may not be really contracts.

201
00:09:38,280 --> 00:09:40,210
And that's why Larry's
going to speak to it.

202
00:09:40,210 --> 00:09:44,370
So remember, even though Nick
Szabo calls it smart contracts,

203
00:09:44,370 --> 00:09:47,010
they may not be smart, and
they may not be contracts.

204
00:09:50,430 --> 00:09:53,260
So a little bit about
the technical features.

205
00:09:53,260 --> 00:09:56,250
Remember our three
ways we did this.

206
00:09:56,250 --> 00:10:00,150
What are the big technical
features that we studied?

207
00:10:00,150 --> 00:10:01,270
I'm sorry to do this.

208
00:10:01,270 --> 00:10:03,840
Anton, what are the
three big buckets?

209
00:10:03,840 --> 00:10:07,200
Remember, we took three
classes and three buckets

210
00:10:07,200 --> 00:10:08,070
of information.

211
00:10:10,794 --> 00:10:12,525
It can be one to three.

212
00:10:12,525 --> 00:10:13,490
AUDIENCE: Cryptography.

213
00:10:13,490 --> 00:10:14,490
PROFESSOR: Cryptography.

214
00:10:14,490 --> 00:10:14,990
Great.

215
00:10:14,990 --> 00:10:15,810
You got one.

216
00:10:15,810 --> 00:10:18,000
Two other people will
give me the others.

217
00:10:18,000 --> 00:10:19,890
So cryptography.

218
00:10:19,890 --> 00:10:20,430
Guess what?

219
00:10:20,430 --> 00:10:25,090
Bitcoin and Ethereum all
have the same cryptography.

220
00:10:25,090 --> 00:10:26,760
It's not identical.

221
00:10:26,760 --> 00:10:29,970
But for the purposes
of design features,

222
00:10:29,970 --> 00:10:33,510
it has cryptographic hashes,
timestamps, block headers,

223
00:10:33,510 --> 00:10:35,760
Merkle trees.

224
00:10:35,760 --> 00:10:38,760
Though Ethereum has more than
one Merkle tree, and bitcoin

225
00:10:38,760 --> 00:10:41,460
has one.

226
00:10:41,460 --> 00:10:44,250
And it has digital
signatures and addresses.

227
00:10:44,250 --> 00:10:47,240
So everything we
talked about three

228
00:10:47,240 --> 00:10:51,950
lectures ago about cryptography,
both forms of blockchains.

229
00:10:51,950 --> 00:10:55,730
Anybody want to give me
what our second bucket was?

230
00:10:55,730 --> 00:10:57,950
Geramo.

231
00:10:57,950 --> 00:10:59,310
AUDIENCE: Decentralization.

232
00:10:59,310 --> 00:11:00,590
PROFESSOR: Decentralization.

233
00:11:00,590 --> 00:11:02,570
What else did we talk about?

234
00:11:02,570 --> 00:11:03,070
Eilon.

235
00:11:03,070 --> 00:11:03,903
AUDIENCE: Consensus.

236
00:11:03,903 --> 00:11:05,250
PROFESSOR: Consensus.

237
00:11:05,250 --> 00:11:10,310
So again, decentralized
network consensus.

238
00:11:10,310 --> 00:11:12,740
Ethereum actually
uses proof of work

239
00:11:12,740 --> 00:11:15,770
even there's some
talk that Ether

240
00:11:15,770 --> 00:11:17,870
will move to proof of stake.

241
00:11:17,870 --> 00:11:20,030
But it's currently
proof of work.

242
00:11:20,030 --> 00:11:22,280
There is a native currency.

243
00:11:22,280 --> 00:11:29,750
It's ETH, or "eh-th," or E-T-H,
or Ether, instead of bitcoin.

244
00:11:29,750 --> 00:11:32,220
And it has a network.

245
00:11:32,220 --> 00:11:36,080
The third thing that
we talked about, Alpha.

246
00:11:36,080 --> 00:11:37,520
AUDIENCE: Transaction format.

247
00:11:37,520 --> 00:11:38,840
PROFESSOR: Transaction format.

248
00:11:38,840 --> 00:11:42,290
So we talked about
transactions and script.

249
00:11:42,290 --> 00:11:43,240
Well, guess what?

250
00:11:43,240 --> 00:11:48,030
Ethereum does not have a
transaction script or UTXO.

251
00:11:48,030 --> 00:11:52,850
So that's the one place
that Vatalik Buterin, who

252
00:11:52,850 --> 00:11:57,020
designed all of this, said
no, had to go a different way.

253
00:11:57,020 --> 00:12:00,070
Instead of transaction
inputs and outputs,

254
00:12:00,070 --> 00:12:02,260
there's something called
state transitions.

255
00:12:05,250 --> 00:12:06,518
Isabella, ledgers.

256
00:12:06,518 --> 00:12:08,310
What are the two
different types of ledgers

257
00:12:08,310 --> 00:12:12,355
that we talked about
and set a class to?

258
00:12:12,355 --> 00:12:14,066
AUDIENCE: Permissioned
and public.

259
00:12:14,066 --> 00:12:17,010
PROFESSOR: So
permissioned and public

260
00:12:17,010 --> 00:12:20,660
are two different
types of blockchains.

261
00:12:20,660 --> 00:12:22,100
Remind me your name again.

262
00:12:22,100 --> 00:12:23,053
Stephanie.

263
00:12:23,053 --> 00:12:25,750
AUDIENCE: Balance ledgers
and transaction ledgers.

264
00:12:25,750 --> 00:12:26,625
PROFESSOR: All right.

265
00:12:26,625 --> 00:12:28,440
So balance ledger and
transaction ledger.

266
00:12:28,440 --> 00:12:30,650
You can keep a list
of transactions.

267
00:12:30,650 --> 00:12:33,638
And this goes back
thousands of years.

268
00:12:33,638 --> 00:12:34,680
This is not a blockchain.

269
00:12:34,680 --> 00:12:37,080
Or you can keep balances.

270
00:12:37,080 --> 00:12:41,080
Bitcoin is, in essence,
a transaction ledger.

271
00:12:41,080 --> 00:12:45,730
Ethereum and many of
the other smart contract

272
00:12:45,730 --> 00:12:48,490
platforms are balance ledgers.

273
00:12:48,490 --> 00:12:53,500
There's a lot of technological
and mathematical reasons why,

274
00:12:53,500 --> 00:12:57,160
which I'd be glad to
do in office hours.

275
00:12:57,160 --> 00:13:00,220
But because of that,
when you're moving

276
00:13:00,220 --> 00:13:03,870
from one set of balances, like--

277
00:13:03,870 --> 00:13:05,170
is it Bo?

278
00:13:05,170 --> 00:13:11,260
Bo has $100 at Bank of
America to Bo having $101.

279
00:13:11,260 --> 00:13:13,150
You need to have a transition.

280
00:13:13,150 --> 00:13:17,050
It's called a state transition,
which would mean add $1.

281
00:13:20,680 --> 00:13:24,580
So Ethereum is account based
and, in fact, doesn't even

282
00:13:24,580 --> 00:13:26,500
have one programming language.

283
00:13:26,500 --> 00:13:28,840
There's six or seven
programming languages

284
00:13:28,840 --> 00:13:32,970
you can write in, for
those who are so inclined.

285
00:13:32,970 --> 00:13:38,570
So I'm going to go through
really quickly my analysis,

286
00:13:38,570 --> 00:13:40,370
and many other
people's, but my sort

287
00:13:40,370 --> 00:13:43,820
of rendition of the difference
between Ethereum and bitcoin.

288
00:13:43,820 --> 00:13:45,380
But stop me if
there's a question.

289
00:13:45,380 --> 00:13:46,750
So the founder.

290
00:13:46,750 --> 00:13:49,400
There's actually a founder.

291
00:13:49,400 --> 00:13:52,450
The mythology around
Vatalik Buterin

292
00:13:52,450 --> 00:13:54,870
might not be as great
as Satoshi Nakamoto.

293
00:13:54,870 --> 00:14:00,500
But a 19-year-old journalist who
was writing about bitcoin for--

294
00:14:00,500 --> 00:14:04,850
was it Bitcoin Magazine, if I
remember where he was writing--

295
00:14:04,850 --> 00:14:08,190
said, we can build
something on top

296
00:14:08,190 --> 00:14:11,630
of it using Nick Szabo's
thoughts about smart contracts,

297
00:14:11,630 --> 00:14:15,135
about six years after bitcoin.

298
00:14:18,150 --> 00:14:19,860
He made it Turing complete.

299
00:14:19,860 --> 00:14:23,670
Does anyone want to remind the
class what Turing complete is?

300
00:14:23,670 --> 00:14:24,720
Did I see a hand up?

301
00:14:24,720 --> 00:14:25,890
Yes.

302
00:14:25,890 --> 00:14:26,455
Bold of you.

303
00:14:26,455 --> 00:14:27,663
What's your first name again?

304
00:14:27,663 --> 00:14:28,390
AUDIENCE: Alana.

305
00:14:28,390 --> 00:14:29,448
PROFESSOR: Alana.

306
00:14:29,448 --> 00:14:31,490
AUDIENCE: You could write
pretty much any-- well,

307
00:14:31,490 --> 00:14:33,782
you can write loops, but you
write any kind of program,

308
00:14:33,782 --> 00:14:35,890
like any form of
logic is expressed.

309
00:14:35,890 --> 00:14:37,800
PROFESSOR: So Alana
says Turing complete

310
00:14:37,800 --> 00:14:39,150
means you can write loops.

311
00:14:39,150 --> 00:14:43,440
But what that really means
for the business crowd is it

312
00:14:43,440 --> 00:14:46,260
means you can write a program
to do just about anything.

313
00:14:46,260 --> 00:14:48,180
It's highly flexible.

314
00:14:48,180 --> 00:14:50,700
I'm sure the technologist in
here will say, well, maybe

315
00:14:50,700 --> 00:14:52,710
there's something
you can't do with it.

316
00:14:52,710 --> 00:14:55,170
But the point in
what Vatalik said

317
00:14:55,170 --> 00:14:58,920
is we shouldn't just limit it
to a scripting language that

318
00:14:58,920 --> 00:15:01,500
was non-Turing
complete that can just

319
00:15:01,500 --> 00:15:05,310
move a little bit of
peer-to-peer value around,

320
00:15:05,310 --> 00:15:11,100
that he wanted to do basically
a peer-to-peer virtual computer

321
00:15:11,100 --> 00:15:14,310
that could move code
and complete code.

322
00:15:14,310 --> 00:15:16,980
And over the years,
it's being able to be

323
00:15:16,980 --> 00:15:18,320
written in a bunch of things.

324
00:15:18,320 --> 00:15:21,810
But Solidity is the
main program language.

325
00:15:21,810 --> 00:15:25,170
And thus, it's account based
rather than transaction

326
00:15:25,170 --> 00:15:27,510
based because of the loops.

327
00:15:27,510 --> 00:15:31,620
If you could loop and be Turing
complete with transactions,

328
00:15:31,620 --> 00:15:34,770
there is a vulnerability
in attack factors

329
00:15:34,770 --> 00:15:38,320
that he made it account based.

330
00:15:38,320 --> 00:15:42,015
The transactions are
stored in Merkle keys.

331
00:15:42,015 --> 00:15:42,640
But guess what?

332
00:15:42,640 --> 00:15:45,220
If you were really interested
and wanted to learn everything

333
00:15:45,220 --> 00:15:48,670
about Ethereum, there's at
least four key Merkle trees

334
00:15:48,670 --> 00:15:52,780
that get summarized up
into the block headers.

335
00:15:52,780 --> 00:15:55,420
There's the transactions,
which, in essence,

336
00:15:55,420 --> 00:15:56,500
aren't transactions.

337
00:15:56,500 --> 00:15:58,780
They call them
state transitions.

338
00:15:58,780 --> 00:16:01,480
There's the state
itself, something

339
00:16:01,480 --> 00:16:04,420
called storage, which is
really related to each one

340
00:16:04,420 --> 00:16:07,390
of the individual Ethereum
accounts, and then,

341
00:16:07,390 --> 00:16:09,670
believe it or not,
actually, receipts.

342
00:16:09,670 --> 00:16:12,700
There's written records of
receipts from every state

343
00:16:12,700 --> 00:16:14,530
transition.

344
00:16:14,530 --> 00:16:19,400
So there's a lot more complexity
inside the Ethereum blockchain.

345
00:16:19,400 --> 00:16:23,510
They run about 14 seconds a
block instead of 10 minutes.

346
00:16:23,510 --> 00:16:26,600
And that change is a
bunch of the economics.

347
00:16:26,600 --> 00:16:32,010
Proof of work, for some
reason, which others can study.

348
00:16:32,010 --> 00:16:35,450
Vatalik decided to use a
different hash function.

349
00:16:35,450 --> 00:16:37,760
It's still based on
elliptic curves, I think,

350
00:16:37,760 --> 00:16:40,840
but a different hash function.

351
00:16:40,840 --> 00:16:42,760
What about the economics?

352
00:16:42,760 --> 00:16:47,680
Well, it's called ETH or E-T-H.
It's programmed in such a way

353
00:16:47,680 --> 00:16:52,210
that you can't use
ASICs, purposely

354
00:16:52,210 --> 00:16:55,840
makes it that mining
is more decentralized,

355
00:16:55,840 --> 00:16:57,580
and you don't have
the big factories.

356
00:17:00,280 --> 00:17:04,360
The entire hash community,
the entire mining community,

357
00:17:04,360 --> 00:17:05,770
is much smaller.

358
00:17:05,770 --> 00:17:06,988
In fact, I think it's about--

359
00:17:06,988 --> 00:17:07,780
what would that be?

360
00:17:07,780 --> 00:17:10,270
200,000 times smaller.

361
00:17:10,270 --> 00:17:15,280
There's 260 terahashes per
second as of yesterday.

362
00:17:15,280 --> 00:17:21,230
And the bitcoin mining
community has 54 exahashes.

363
00:17:21,230 --> 00:17:23,900
That means lots more computers,
a lot more electricity,

364
00:17:23,900 --> 00:17:28,490
a lot more gnashing of
teeth about resources.

365
00:17:28,490 --> 00:17:32,060
Bitcoin started in
January of 2009.

366
00:17:32,060 --> 00:17:36,920
And nobody owned any bitcoin,
whereas Ethereum started

367
00:17:36,920 --> 00:17:40,955
with a presale and an ICO,
which we'll be talking about,

368
00:17:40,955 --> 00:17:44,530
about 72 million Ethereum.

369
00:17:44,530 --> 00:17:47,950
Vatalik wanted to raise money
he was maybe 19 years old.

370
00:17:47,950 --> 00:17:49,870
He looped in with a
venture capitalist

371
00:17:49,870 --> 00:17:54,460
from Canada, Joe Lubin,
who now runs ConsenSys.

372
00:17:54,460 --> 00:17:59,640
Lubin took about 10% or 9
and 1/2% of the offering.

373
00:17:59,640 --> 00:18:01,990
They put 9 and 1/2%
in a foundation

374
00:18:01,990 --> 00:18:04,720
called the Ethereum Foundation.

375
00:18:04,720 --> 00:18:07,930
And the other 80% was sold to
the public for $18 million.

376
00:18:10,732 --> 00:18:12,670
We'll talk later
in the semester as

377
00:18:12,670 --> 00:18:15,100
to whether that was really
a securities offering.

378
00:18:15,100 --> 00:18:20,110
I've publicly said I think
so, but that was in 2014.

379
00:18:20,110 --> 00:18:23,170
And in 2018, the Securities
and Exchange Commission

380
00:18:23,170 --> 00:18:26,680
has said regardless of why
it might have been in '14,

381
00:18:26,680 --> 00:18:28,840
it's now sufficiently
decentralized

382
00:18:28,840 --> 00:18:31,300
that we'll consider
it not a security.

383
00:18:31,300 --> 00:18:33,280
But in essence, they
raised $18 million

384
00:18:33,280 --> 00:18:37,046
and were off to the races.

385
00:18:37,046 --> 00:18:41,810
Oh, and the 10% that Joe Lubin
kept, if he still owned it,

386
00:18:41,810 --> 00:18:44,100
would be worth about
$2 billion now.

387
00:18:44,100 --> 00:18:48,180
But he probably sold
some of it along the way.

388
00:18:48,180 --> 00:18:50,090
There are block rewards.

389
00:18:50,090 --> 00:18:53,690
The monetary policy in
Ethereum, if you remember,

390
00:18:53,690 --> 00:18:56,900
it splits every four years.

391
00:18:56,900 --> 00:18:58,570
There's the block reward splits.

392
00:18:58,570 --> 00:19:03,860
Ethereum was set up that it
was five Ether per block.

393
00:19:03,860 --> 00:19:06,050
And then a year ago,
they just announced

394
00:19:06,050 --> 00:19:07,760
they were going to
change the software

395
00:19:07,760 --> 00:19:10,040
to three Ether per block.

396
00:19:10,040 --> 00:19:11,720
And a month ago, they
announced they're

397
00:19:11,720 --> 00:19:13,760
going to change it to
two Ether per block.

398
00:19:13,760 --> 00:19:18,500
So the monetary policy of
bitcoin is said to be immutable

399
00:19:18,500 --> 00:19:20,210
and fixed forever.

400
00:19:20,210 --> 00:19:23,750
But I would say that because
Ethereum supposedly was fixed

401
00:19:23,750 --> 00:19:27,800
and now, twice, they've
changed the monetary policy,

402
00:19:27,800 --> 00:19:32,240
there is a form that if all
the miners and all the computer

403
00:19:32,240 --> 00:19:35,420
nodes decide, they can
change the monetary policy.

404
00:19:35,420 --> 00:19:39,050
And Ethereum has shown that.

405
00:19:39,050 --> 00:19:41,750
I think Ethereum's a
bit more centralized

406
00:19:41,750 --> 00:19:44,990
and has more leadership,
because Vatalik Buterin

407
00:19:44,990 --> 00:19:49,820
is an actual human who is
willing to disclose who he is.

408
00:19:49,820 --> 00:19:52,730
And he still has the sort
of founder following.

409
00:19:52,730 --> 00:19:54,650
And Satoshi Nakamoto went--

410
00:19:54,650 --> 00:19:57,440
he ghosted all of
us, in a sense.

411
00:19:57,440 --> 00:20:00,910
And so thus, it's in some
ways socially just more

412
00:20:00,910 --> 00:20:04,050
decentralized.

413
00:20:04,050 --> 00:20:07,950
And then fees are
voluntary in bitcoin.

414
00:20:07,950 --> 00:20:12,880
They're a necessary part
to channel everything.

415
00:20:12,880 --> 00:20:13,380
Emily.

416
00:20:13,380 --> 00:20:15,150
AUDIENCE: Going back
to monetary policy,

417
00:20:15,150 --> 00:20:17,360
in practice, what
does it actually mean

418
00:20:17,360 --> 00:20:20,850
for a difference between bitcoin
and Ethereum that it's fixed

419
00:20:20,850 --> 00:20:23,440
but changing versus
staying at the same rate?

420
00:20:23,440 --> 00:20:25,860
PROFESSOR: So Emily asks,
what does it mean in practice?

421
00:20:25,860 --> 00:20:33,300
In practice, bitcoin has a much
slower growth rate or inflation

422
00:20:33,300 --> 00:20:40,230
rate, which is currently, if
I'm not mistaken, about half.

423
00:20:40,230 --> 00:20:41,130
What's that?

424
00:20:41,130 --> 00:20:42,190
About 4%.

425
00:20:42,190 --> 00:20:46,590
But it's about half of the
inflation rate at Ethereum,

426
00:20:46,590 --> 00:20:49,830
which is running
in the 7% range.

427
00:20:49,830 --> 00:20:53,670
Every one of the
1,600 live tokens

428
00:20:53,670 --> 00:20:55,500
that are right
now Ott coins have

429
00:20:55,500 --> 00:20:57,210
separate monetary policies.

430
00:20:57,210 --> 00:20:59,220
And one could do a
complete economic study

431
00:20:59,220 --> 00:21:00,660
about what it means.

432
00:21:00,660 --> 00:21:02,970
But in terms of
Ethereum versus bitcoin,

433
00:21:02,970 --> 00:21:07,140
Ethereum has a higher
inflation rate.

434
00:21:07,140 --> 00:21:09,270
And secondly,
because they've shown

435
00:21:09,270 --> 00:21:13,650
that they can form some
consensus and change it,

436
00:21:13,650 --> 00:21:15,780
I think it sends an
interesting question

437
00:21:15,780 --> 00:21:21,780
into that ecosystem as to how
hard a monetary policy versus

438
00:21:21,780 --> 00:21:27,130
a more human-based,
flexible monetary policy.

439
00:21:27,130 --> 00:21:28,180
Hugo.

440
00:21:28,180 --> 00:21:31,840
AUDIENCE: So is it that the
core developers of the Ethereum

441
00:21:31,840 --> 00:21:33,822
blockchain decide?

442
00:21:33,822 --> 00:21:36,280
I'm not familiar with how they
make that decision to change

443
00:21:36,280 --> 00:21:37,520
the monetary policy.

444
00:21:37,520 --> 00:21:41,350
PROFESSOR: So the core
developers of any coin

445
00:21:41,350 --> 00:21:44,560
can propose to the
various mining and nodes

446
00:21:44,560 --> 00:21:50,510
to make a change, including
the monetary policy.

447
00:21:50,510 --> 00:21:53,290
Even in bitcoin,
the monetary policy

448
00:21:53,290 --> 00:21:56,230
could change if
there was a proposal

449
00:21:56,230 --> 00:21:58,300
from the core developers.

450
00:21:58,300 --> 00:22:01,480
In the Ethereum case,
they have done that twice.

451
00:22:01,480 --> 00:22:03,160
The core development
of Ethereum is

452
00:22:03,160 --> 00:22:07,720
highly concentrated around
the Ethereum Foundation.

453
00:22:07,720 --> 00:22:11,110
ConsenSys, the company
that Joe Lubin runs,

454
00:22:11,110 --> 00:22:15,970
has some normative social
standing in the community,

455
00:22:15,970 --> 00:22:18,070
as well.

456
00:22:18,070 --> 00:22:20,650
But it's been part of proposals.

457
00:22:20,650 --> 00:22:25,780
And in reading the proposals, it
goes back to Emily's question.

458
00:22:25,780 --> 00:22:28,540
The core developers have said we
should bring down the inflation

459
00:22:28,540 --> 00:22:29,350
rate.

460
00:22:29,350 --> 00:22:32,570
There's an active advocacy to
bring down the inflation rate.

461
00:22:32,570 --> 00:22:35,000
AUDIENCE: So you go by
updating your node software?

462
00:22:35,000 --> 00:22:36,790
PROFESSOR: Yes, in essence.

463
00:22:36,790 --> 00:22:39,880
And it could lead to a
hard fork, I believe.

464
00:22:42,588 --> 00:22:43,630
So that's the background.

465
00:22:43,630 --> 00:22:45,172
Let me just talk
about the platforms.

466
00:22:45,172 --> 00:22:47,230
We know about Ethereum.

467
00:22:47,230 --> 00:22:48,550
We've just chatted about it.

468
00:22:48,550 --> 00:22:51,910
Its actual current market
value is $22 billion.

469
00:22:51,910 --> 00:22:54,490
While we're not going to spend
a lot of time in this class

470
00:22:54,490 --> 00:22:56,080
this semester about
market values,

471
00:22:56,080 --> 00:22:57,538
I thought it would
give you a sense

472
00:22:57,538 --> 00:22:58,870
of how people think about it.

473
00:22:58,870 --> 00:23:02,240
The other five that you
did some reading about.

474
00:23:02,240 --> 00:23:05,260
EOS, whose market value
is about $5 billion,

475
00:23:05,260 --> 00:23:09,070
is so new, because they just
finished an initial coin

476
00:23:09,070 --> 00:23:10,900
offering in July.

477
00:23:10,900 --> 00:23:12,770
And they just went live in July.

478
00:23:12,770 --> 00:23:14,950
But they raised $4.2 billion.

479
00:23:14,950 --> 00:23:19,900
So my estimate, my prediction,
is EOS is going to be real

480
00:23:19,900 --> 00:23:22,030
and going to be around.

481
00:23:22,030 --> 00:23:25,360
They've taken $1 billion
of that $4.2 billion

482
00:23:25,360 --> 00:23:31,370
and set up a Kickstarter
sort of venture firm

483
00:23:31,370 --> 00:23:35,570
literally so that they could
maybe support some of you.

484
00:23:35,570 --> 00:23:39,110
You could be knocking on
the EOS venture firm's door

485
00:23:39,110 --> 00:23:40,610
and saying fund me.

486
00:23:40,610 --> 00:23:41,750
I've got a great idea.

487
00:23:41,750 --> 00:23:44,450
And they'd have one
condition, amongst others.

488
00:23:44,450 --> 00:23:47,570
You'd have to use
EOS as the platform.

489
00:23:47,570 --> 00:23:51,440
So it's a self-reinforcing
business model.

490
00:23:51,440 --> 00:23:54,320
NEO is the other big
one that I'd mention,

491
00:23:54,320 --> 00:23:58,640
which was started two years ago
out of China, a lot of people

492
00:23:58,640 --> 00:24:01,280
think with the backing of
the Chinese government.

493
00:24:01,280 --> 00:24:05,000
Even if not officially,
certainly with the verbal

494
00:24:05,000 --> 00:24:08,900
and help of there.

495
00:24:08,900 --> 00:24:11,440
Some people think of it
as the Ethereum of China,

496
00:24:11,440 --> 00:24:14,420
but it does use a
different proof of work.

497
00:24:14,420 --> 00:24:15,670
And those are the main ones.

498
00:24:15,670 --> 00:24:17,870
Ethereum Classic is
there because there

499
00:24:17,870 --> 00:24:25,660
was a hard fork off of Ethereum
after the DAO circumstance.

500
00:24:25,660 --> 00:24:29,490
Any quick questions
about the platforms?

501
00:24:29,490 --> 00:24:32,910
And NEO and EOS
are thought to be

502
00:24:32,910 --> 00:24:36,030
a little bit higher
throughput and faster because

503
00:24:36,030 --> 00:24:39,720
of the different proof of
work that each of them have.

504
00:24:39,720 --> 00:24:43,626
So maybe they're more scalable
than Ethereum long term.

505
00:24:43,626 --> 00:24:44,792
AUDIENCE: [INAUDIBLE]

506
00:24:44,792 --> 00:24:45,750
PROFESSOR: What's that?

507
00:24:45,750 --> 00:24:47,340
AUDIENCE: [INAUDIBLE]

508
00:24:47,340 --> 00:24:49,810
PROFESSOR: So Hugo is pointing
out that neither one truly

509
00:24:49,810 --> 00:24:50,310
use--

510
00:24:50,310 --> 00:24:55,710
NEO and EOS truly use
decentralized proof of work.

511
00:24:55,710 --> 00:25:00,630
They use a delegated
Byzantine fault tolerance.

512
00:25:00,630 --> 00:25:01,800
This was from the reading.

513
00:25:01,800 --> 00:25:03,300
We're not going to
spend time on it.

514
00:25:03,300 --> 00:25:08,730
But there was 12 use cases that
the Digital Chamber of Commerce

515
00:25:08,730 --> 00:25:09,322
came up with.

516
00:25:09,322 --> 00:25:11,280
And we're going to study
most of these, not all

517
00:25:11,280 --> 00:25:12,613
of these, later in the semester.

518
00:25:15,390 --> 00:25:17,700
Please tell me, Sean.

519
00:25:17,700 --> 00:25:19,890
AUDIENCE: So from the
growth perspective

520
00:25:19,890 --> 00:25:22,680
across different
platforms, do you

521
00:25:22,680 --> 00:25:27,520
support multi programming
language platforms as opposed

522
00:25:27,520 --> 00:25:31,962
to single-language
programming platforms?

523
00:25:31,962 --> 00:25:35,700
Which one do you think
has more potential?

524
00:25:35,700 --> 00:25:40,690
PROFESSOR: The question is,
do multi-language platforms

525
00:25:40,690 --> 00:25:44,230
have more potential or less
than single-language platforms?

526
00:25:44,230 --> 00:25:45,660
I haven't done a real study.

527
00:25:45,660 --> 00:25:49,190
My gut tells me that
there's some trade-offs.

528
00:25:49,190 --> 00:25:51,010
So a multi-language
platform probably

529
00:25:51,010 --> 00:25:53,620
allows for more
development of apps,

530
00:25:53,620 --> 00:25:56,170
just like on the
internet, if you

531
00:25:56,170 --> 00:26:00,040
can have more app development.

532
00:26:00,040 --> 00:26:02,770
Though if you have only
one single language,

533
00:26:02,770 --> 00:26:06,490
you might have fewer
vulnerabilities,

534
00:26:06,490 --> 00:26:10,060
vulnerabilities either
to attack or coming down.

535
00:26:10,060 --> 00:26:14,050
So there's probably
some trade-offs.

536
00:26:14,050 --> 00:26:19,870
And I'm more into
democratizing capital markets,

537
00:26:19,870 --> 00:26:22,550
so probably my biases
would be to the ones that

538
00:26:22,550 --> 00:26:24,560
have multi-language potential.

539
00:26:24,560 --> 00:26:27,830
But I can't say for sure
whether it will necessarily

540
00:26:27,830 --> 00:26:32,330
win, because there are some
trade-offs, if that helps.

541
00:26:32,330 --> 00:26:36,170
So we'll take a view of all of
these later in the semester.

542
00:26:36,170 --> 00:26:39,790
But what's interesting is none
of them have yet taken off.

543
00:26:39,790 --> 00:26:45,440
And in fact, if you
look at the dApps today,

544
00:26:45,440 --> 00:26:47,690
the decentralized
applications which

545
00:26:47,690 --> 00:26:51,080
run on a decentralized
blockchain,

546
00:26:51,080 --> 00:26:53,510
they generally
have native tokens.

547
00:26:53,510 --> 00:26:55,880
They don't have to have a
native token, by the way.

548
00:26:55,880 --> 00:27:00,380
You can run a dApp
based on Ether.

549
00:27:00,380 --> 00:27:02,930
You could run a dApp
on an underlying token.

550
00:27:02,930 --> 00:27:07,260
But they almost always have
their own native token.

551
00:27:07,260 --> 00:27:10,200
This is a list that I pulled
off the internet yesterday

552
00:27:10,200 --> 00:27:11,550
similar to your reading.

553
00:27:11,550 --> 00:27:15,090
These are the actual dApps
that are working the highest--

554
00:27:15,090 --> 00:27:19,520
gaming, gambling,
exchanges, and finance.

555
00:27:19,520 --> 00:27:25,200
The most active dApp
is a gambling site that

556
00:27:25,200 --> 00:27:28,050
only has 1,500 users a day.

557
00:27:28,050 --> 00:27:29,550
In the last 24 hours--

558
00:27:29,550 --> 00:27:33,000
I pulled this down yesterday,
so that was a Monday.

559
00:27:33,000 --> 00:27:35,670
So this would have been
Sunday night to Monday night.

560
00:27:35,670 --> 00:27:40,980
There was only 1,500 users to
the most active gambling site.

561
00:27:40,980 --> 00:27:42,960
CryptoKitties that
some of you might know

562
00:27:42,960 --> 00:27:46,890
had 418 users in those 24 hours.

563
00:27:46,890 --> 00:27:53,360
These are not economy
wise use cases.

564
00:27:53,360 --> 00:27:55,810
You would see numbers that
were hundreds of thousands.

565
00:27:55,810 --> 00:27:59,440
Or if it's like Facebook,
there's two billion members.

566
00:27:59,440 --> 00:28:02,140
And probably in any 24 hours,
there's a half a billion

567
00:28:02,140 --> 00:28:04,390
to a billion people that
check in on Facebook.

568
00:28:04,390 --> 00:28:07,610
So these are pretty
limited at this stage.

569
00:28:07,610 --> 00:28:09,718
But it has led to something
called initial coin

570
00:28:09,718 --> 00:28:11,260
offerings, which
we're going to study

571
00:28:11,260 --> 00:28:13,270
a bunch later in the semester.

572
00:28:13,270 --> 00:28:17,470
Initial coin offerings raise
money to build a network.

573
00:28:17,470 --> 00:28:22,290
But the tokens are usually
issued before they're usable.

574
00:28:22,290 --> 00:28:24,430
And a token is some
native currency

575
00:28:24,430 --> 00:28:27,620
to be used on a network.

576
00:28:27,620 --> 00:28:32,160
The development is open
but highly centralized.

577
00:28:32,160 --> 00:28:35,310
The promoters usually
give themselves some skin

578
00:28:35,310 --> 00:28:38,130
in the game or some vig.

579
00:28:38,130 --> 00:28:41,430
In the Ethereum
initial coin offering,

580
00:28:41,430 --> 00:28:43,180
the Foundation kept 10%.

581
00:28:43,180 --> 00:28:44,580
Joe Lubin kept 10%.

582
00:28:44,580 --> 00:28:46,950
And the other 80% was
sold to the public.

583
00:28:46,950 --> 00:28:52,290
But some keep 60% or
80% for themselves.

584
00:28:52,290 --> 00:28:56,160
I mean, it's a whole wide
range of economic incentives

585
00:28:56,160 --> 00:28:57,940
and models.

586
00:28:57,940 --> 00:29:01,000
And the tokens are usually
fungible and transferable.

587
00:29:01,000 --> 00:29:03,820
And thus, you can sort
put them on an exchange

588
00:29:03,820 --> 00:29:07,000
and try to sell them
and promote them.

589
00:29:07,000 --> 00:29:08,500
And again, we're
going to come back.

590
00:29:08,500 --> 00:29:10,720
But I thought if we're
talking about smart contracts,

591
00:29:10,720 --> 00:29:14,740
they tie into initial
coin offerings.

592
00:29:14,740 --> 00:29:16,540
Daniel.

593
00:29:16,540 --> 00:29:18,710
AUDIENCE: If the tokens
aren't functional,

594
00:29:18,710 --> 00:29:21,342
what goes into their evaluation?

595
00:29:21,342 --> 00:29:23,050
PROFESSOR: That is an
excellent question.

596
00:29:23,050 --> 00:29:25,750
Daniel says, what goes into
the valuation of something

597
00:29:25,750 --> 00:29:28,050
that is not functional?

598
00:29:28,050 --> 00:29:31,810
And there are many
things that go in.

599
00:29:31,810 --> 00:29:34,900
Ultimately, it's
in the anticipation

600
00:29:34,900 --> 00:29:39,010
of the potential profit and
appreciation in the future.

601
00:29:39,010 --> 00:29:41,770
So if you were just
to have a laundromat,

602
00:29:41,770 --> 00:29:44,560
a corner laundromat
here in Cambridge,

603
00:29:44,560 --> 00:29:47,650
you probably would only pay
for the laundromat token what

604
00:29:47,650 --> 00:29:53,920
you thought the value was to do
your laundry, $0.75 or whatever

605
00:29:53,920 --> 00:29:55,220
the--

606
00:29:55,220 --> 00:29:59,290
frankly, I don't know what the
tokens go for in Cambridge.

607
00:29:59,290 --> 00:30:02,410
You're going to tell me what
a laundromat token goes for?

608
00:30:02,410 --> 00:30:05,650
But if it's prefunctional
and the laundromat has not

609
00:30:05,650 --> 00:30:11,190
been built yet and you have
confidence it will be built,

610
00:30:11,190 --> 00:30:12,440
you'll probably discount back.

611
00:30:12,440 --> 00:30:17,845
You might pay $0.50 instead
of the $0.75, ultimately.

612
00:30:17,845 --> 00:30:20,220
If you think it's going to be
a really popular laundromat

613
00:30:20,220 --> 00:30:24,360
and it's going to be the place
to be seen by all MIT students

614
00:30:24,360 --> 00:30:28,270
and you think that token
might one day be worth $5,

615
00:30:28,270 --> 00:30:29,670
then you're starting
to speculate

616
00:30:29,670 --> 00:30:32,710
on the community's interest.

617
00:30:32,710 --> 00:30:37,660
So these tokens, if you
really believe in them,

618
00:30:37,660 --> 00:30:39,480
you're trying to pay
for what others will

619
00:30:39,480 --> 00:30:41,280
pay for them in the future.

620
00:30:41,280 --> 00:30:43,830
And the economic equilibrium
would be discounted.

621
00:30:43,830 --> 00:30:45,720
So thus, you're buying
it because you think

622
00:30:45,720 --> 00:30:48,357
they'll appreciate in value.

623
00:30:48,357 --> 00:30:50,190
AUDIENCE: Couldn't you
make an argument sort

624
00:30:50,190 --> 00:30:52,710
of following
numismatics, how people

625
00:30:52,710 --> 00:30:56,320
assign some aesthetic
value to physical coins?

626
00:30:56,320 --> 00:30:58,890
So, too, you could
have an impulse

627
00:30:58,890 --> 00:31:00,182
beyond speculation [INAUDIBLE].

628
00:31:00,182 --> 00:31:01,432
PROFESSOR: Your first name is?

629
00:31:01,432 --> 00:31:02,130
AUDIENCE: Isaac.

630
00:31:02,130 --> 00:31:02,838
PROFESSOR: Isaac.

631
00:31:02,838 --> 00:31:08,400
Isaac says maybe there could
be a numismatic or other value

632
00:31:08,400 --> 00:31:09,090
to it.

633
00:31:09,090 --> 00:31:10,380
It's possible.

634
00:31:10,380 --> 00:31:13,140
I mean, certainly,
CryptoKitties caught

635
00:31:13,140 --> 00:31:16,620
into the whole
collectibles wave.

636
00:31:16,620 --> 00:31:20,550
Let me try to wrap up so that
I can hand it off to Larry.

637
00:31:20,550 --> 00:31:23,670
So there's been
$28 billion raised

638
00:31:23,670 --> 00:31:26,340
per one website, Elementis.

639
00:31:26,340 --> 00:31:27,930
Others that track
it think it's only

640
00:31:27,930 --> 00:31:30,540
between $20 and $25 billion.

641
00:31:30,540 --> 00:31:34,620
There are no official arithmetic
and no official records.

642
00:31:34,620 --> 00:31:38,880
But the $28 billion raised
in initial coin offerings,

643
00:31:38,880 --> 00:31:41,790
this is a cut of a
neat video that I just

644
00:31:41,790 --> 00:31:44,400
took the last slide
of from Elementis.

645
00:31:44,400 --> 00:31:46,140
You can watch it in
a minute and a half.

646
00:31:46,140 --> 00:31:50,250
And over the years,
it's geographically

647
00:31:50,250 --> 00:31:54,750
dispersed with blue
being Oceana, Australia,

648
00:31:54,750 --> 00:31:59,940
and so forth, green being Asia,
North America being orange.

649
00:31:59,940 --> 00:32:02,730
But the biggest ones,
EOS raised $4.2 billion.

650
00:32:02,730 --> 00:32:06,850
Telegram, $1.7
billion and the like.

651
00:32:06,850 --> 00:32:08,730
These are not small figures.

652
00:32:08,730 --> 00:32:15,860
There's been at least
4,500 proposed initial coin

653
00:32:15,860 --> 00:32:16,400
offerings.

654
00:32:16,400 --> 00:32:20,900
Many of them didn't raise any
money, but over 2,000 to 3,000

655
00:32:20,900 --> 00:32:22,670
have raised money.

656
00:32:22,670 --> 00:32:24,290
I'm going to use
the 3,000 figure.

657
00:32:24,290 --> 00:32:29,110
Less than half of them
are even live today.

658
00:32:29,110 --> 00:32:32,110
There's one study that
shows that 59% fail

659
00:32:32,110 --> 00:32:35,070
within the first nine months.

660
00:32:35,070 --> 00:32:39,510
There are other studies that say
between 1/4 and 3/4 are scams.

661
00:32:39,510 --> 00:32:42,690
Christian Catalina
here has a study

662
00:32:42,690 --> 00:32:46,060
which I think is probably
the most valid and reliable,

663
00:32:46,060 --> 00:32:48,720
which says 25% are
scams or frauds.

664
00:32:48,720 --> 00:32:52,746
Statis has one that
says 80% might be.

665
00:32:52,746 --> 00:32:54,347
AUDIENCE: You said 59% fail.

666
00:32:54,347 --> 00:32:56,430
Can you talk more about
what "failure" means here?

667
00:32:56,430 --> 00:32:58,222
Does that mean it just
goes to 0 or there's

668
00:32:58,222 --> 00:33:01,057
a technological deficiency?

669
00:33:01,057 --> 00:33:02,640
PROFESSOR: Can I
hold that until later

670
00:33:02,640 --> 00:33:06,480
when we're going to dig
into these in the class?

671
00:33:06,480 --> 00:33:10,710
But broad definition of fail,
so it's not just technology.

672
00:33:10,710 --> 00:33:14,400
It means that you can't even
find the website anymore.

673
00:33:14,400 --> 00:33:17,080
They took your
money, and they ran.

674
00:33:17,080 --> 00:33:23,520
Or they still have a
website, but it's not live.

675
00:33:23,520 --> 00:33:26,490
It doesn't seem like there's
a development team any longer.

676
00:33:26,490 --> 00:33:29,050
So multiple
definitions of failure.

677
00:33:29,050 --> 00:33:30,750
So that brings us
to legal issues.

678
00:33:30,750 --> 00:33:33,580
And as Larry brings
up his computer,

679
00:33:33,580 --> 00:33:37,090
let me just introduce our guest
lecturer, Larry Lessig, who's

680
00:33:37,090 --> 00:33:40,220
an esteemed Harvard
professor of law.

681
00:33:40,220 --> 00:33:41,140
He was at Stanford.

682
00:33:41,140 --> 00:33:44,380
He started something
called the Center

683
00:33:44,380 --> 00:33:46,690
for Internet and Society.

684
00:33:46,690 --> 00:33:50,290
You clerked for
Justice Scalia, so you

685
00:33:50,290 --> 00:33:53,440
must have some real views on
what's going on this week,

686
00:33:53,440 --> 00:33:55,300
then.

687
00:33:55,300 --> 00:34:01,270
And this incredible appellate
court judge, Posner.

688
00:34:01,270 --> 00:34:04,720
And Larry also-- his first of--

689
00:34:04,720 --> 00:34:06,712
whatever you're up
to, 9 or 10 books now?

690
00:34:06,712 --> 00:34:07,920
LARRY LESSIG: It's something.

691
00:34:07,920 --> 00:34:08,560
I don't know.

692
00:34:08,560 --> 00:34:10,690
PROFESSOR: But his first
was code and the law

693
00:34:10,690 --> 00:34:13,699
that I referenced in
our first lecture.

694
00:34:13,699 --> 00:34:15,150
Larry is going to take it away.

695
00:34:15,150 --> 00:34:16,150
LARRY LESSIG: OK, great.

696
00:34:16,150 --> 00:34:19,120
So I'm really happy
to be able to say

697
00:34:19,120 --> 00:34:22,840
that I've taught business
school students at MIT.

698
00:34:22,840 --> 00:34:26,179
I mean, I'm a tech wannabe.

699
00:34:26,179 --> 00:34:30,429
But I have never been allowed
to be officially here,

700
00:34:30,429 --> 00:34:32,980
so this is exciting.

701
00:34:32,980 --> 00:34:36,400
But what I want to
do is really address

702
00:34:36,400 --> 00:34:40,449
what I've experienced
as misperceptions

703
00:34:40,449 --> 00:34:43,000
about the nature
of the law as it

704
00:34:43,000 --> 00:34:46,360
might interact with the issues
around digital contracts.

705
00:34:46,360 --> 00:34:47,679
So I teach contracts.

706
00:34:47,679 --> 00:34:50,230
Contract law is one of the
subjects which I've taught

707
00:34:50,230 --> 00:34:52,630
since I started teaching.

708
00:34:52,630 --> 00:34:55,600
And so I want to frame
four points for you

709
00:34:55,600 --> 00:34:57,550
about how to think
about contract law

710
00:34:57,550 --> 00:35:02,055
as it relates to these
potential digital contracts.

711
00:35:02,055 --> 00:35:03,430
But the first
thing to do is just

712
00:35:03,430 --> 00:35:04,720
to make sure we
understand we're all

713
00:35:04,720 --> 00:35:06,040
talking about the same thing.

714
00:35:06,040 --> 00:35:06,540
OK.

715
00:35:06,540 --> 00:35:11,320
So here's what the
core of contract law

716
00:35:11,320 --> 00:35:14,640
teaches us contracts are.

717
00:35:14,640 --> 00:35:17,130
Contract is a promise
or performance

718
00:35:17,130 --> 00:35:22,130
given in exchange for a
promise or performance.

719
00:35:22,130 --> 00:35:24,770
"Given in exchange for"
is really critical.

720
00:35:24,770 --> 00:35:26,960
That's the quid pro
quo that's sometimes

721
00:35:26,960 --> 00:35:30,950
referred to as the
consideration for the promise.

722
00:35:30,950 --> 00:35:32,532
But notice here,
this is mapping out

723
00:35:32,532 --> 00:35:33,740
four different possibilities.

724
00:35:33,740 --> 00:35:36,800
You can have a promise in
exchange for a promise.

725
00:35:36,800 --> 00:35:40,780
I promise to pay you $10,000
if you promise to sing

726
00:35:40,780 --> 00:35:43,450
an opera for me tomorrow.

727
00:35:43,450 --> 00:35:46,150
Or it can be a promise
for a performance.

728
00:35:46,150 --> 00:35:52,100
I promise to pay you, Andy, $5
if you sing a song for us now.

729
00:35:52,100 --> 00:35:54,110
So it's the performance I want.

730
00:35:54,110 --> 00:35:56,110
I don't want a promise
from him, because we know

731
00:35:56,110 --> 00:35:58,780
what Andy's promises are worth.

732
00:35:58,780 --> 00:36:00,670
Or it could be a
performance for a promise.

733
00:36:00,670 --> 00:36:05,350
I'll sing if you promise
to pay me after I'm done.

734
00:36:05,350 --> 00:36:07,330
Or the really interesting
one for our purposes

735
00:36:07,330 --> 00:36:11,260
is a performance
for a performance.

736
00:36:11,260 --> 00:36:14,790
I'll sing if you pay me $5,000.

737
00:36:14,790 --> 00:36:17,290
I didn't want a promise from
you, because I don't trust you.

738
00:36:17,290 --> 00:36:18,730
You're business school students.

739
00:36:18,730 --> 00:36:21,790
So I just want the money.

740
00:36:21,790 --> 00:36:23,458
But I'm not promising
to do anything.

741
00:36:23,458 --> 00:36:25,250
I'm actually going to
do something for you.

742
00:36:25,250 --> 00:36:27,370
Now, that's the one
that we're going

743
00:36:27,370 --> 00:36:30,407
to think about in the
context of digital contracts.

744
00:36:30,407 --> 00:36:30,990
Oh, I'm sorry.

745
00:36:30,990 --> 00:36:34,718
Let me just fix one thing here.

746
00:36:34,718 --> 00:36:36,510
Actually, I was going
to make three points,

747
00:36:36,510 --> 00:36:37,880
but then I decided I
was going to make four.

748
00:36:37,880 --> 00:36:38,672
So let me fix that.

749
00:36:38,672 --> 00:36:40,962
Four points about contracts.

750
00:36:40,962 --> 00:36:42,420
So the first point
about contracts,

751
00:36:42,420 --> 00:36:45,035
so think about this case.

752
00:36:45,035 --> 00:36:46,830
All right.

753
00:36:46,830 --> 00:36:49,990
It's not quite digital, but
it's a physical contract

754
00:36:49,990 --> 00:36:55,640
machine in the sense that it's
a performance for a performance.

755
00:36:55,640 --> 00:36:59,630
If you drop a dime
into this machine,

756
00:36:59,630 --> 00:37:04,590
then out will come Dr.
Pepper or something else.

757
00:37:04,590 --> 00:37:05,090
OK.

758
00:37:05,090 --> 00:37:08,960
So there's no promises involved.

759
00:37:08,960 --> 00:37:10,940
And we understand there's
a mechanism that's

760
00:37:10,940 --> 00:37:14,180
to produce this result.
And obviously, these

761
00:37:14,180 --> 00:37:16,860
have been here for a long time.

762
00:37:16,860 --> 00:37:19,070
And these mechanisms
provide real value,

763
00:37:19,070 --> 00:37:21,920
because to the extent you
don't have to hire somebody

764
00:37:21,920 --> 00:37:25,400
to stand there handing
out Dr. Peppers,

765
00:37:25,400 --> 00:37:28,490
you can lower the cost
of delivering Dr. Peppers

766
00:37:28,490 --> 00:37:30,230
and increase the
market for Dr. Pepper.

767
00:37:30,230 --> 00:37:32,450
So this is the
motivation-- lower

768
00:37:32,450 --> 00:37:34,430
the transaction
costs of engaging

769
00:37:34,430 --> 00:37:36,980
in a particular kind
of contract, which

770
00:37:36,980 --> 00:37:38,690
is performance for performance.

771
00:37:38,690 --> 00:37:40,190
Now, when you look
at that contract,

772
00:37:40,190 --> 00:37:44,750
you should ask yourself, what
are the terms of this contract?

773
00:37:44,750 --> 00:37:45,250
All right.

774
00:37:45,250 --> 00:37:47,710
So some of them are express,
and they're pretty obvious.

775
00:37:47,710 --> 00:37:48,830
So this says $0.10.

776
00:37:48,830 --> 00:37:56,580
It says if you pay $0.10,
you will get the Dr. Pepper.

777
00:37:56,580 --> 00:37:58,020
Or that's what you
kind of expect.

778
00:37:58,020 --> 00:38:00,540
There's a great cartoon I
couldn't find where you come

779
00:38:00,540 --> 00:38:02,860
to a machine that
says deposit $0.50.

780
00:38:02,860 --> 00:38:04,920
Deposits $0.50, and
the light comes on.

781
00:38:04,920 --> 00:38:07,690
It says, thank you very much.

782
00:38:07,690 --> 00:38:08,190
Right.

783
00:38:08,190 --> 00:38:10,950
So you think you know
what this contract is,

784
00:38:10,950 --> 00:38:13,915
but that joke hints
that maybe you don't

785
00:38:13,915 --> 00:38:15,040
know what that contract is.

786
00:38:15,040 --> 00:38:17,610
But the express terms
of the one we expect

787
00:38:17,610 --> 00:38:20,370
go with a statement,
the machine.

788
00:38:20,370 --> 00:38:23,630
But then there's a whole
bunch of implied terms.

789
00:38:23,630 --> 00:38:27,310
So one implied term if this
machine is in the United States

790
00:38:27,310 --> 00:38:31,270
is that when you take the Dr.
Pepper out and you drink it,

791
00:38:31,270 --> 00:38:34,000
it's actually going
to be safe to drink.

792
00:38:34,000 --> 00:38:35,770
It's actually going
to be Dr. Pepper.

793
00:38:35,770 --> 00:38:38,360
It's actually not
going to make you sick.

794
00:38:38,360 --> 00:38:42,350
And none of that is written
on the Dr. Pepper machine.

795
00:38:42,350 --> 00:38:45,590
That's instead a
term, a contract term,

796
00:38:45,590 --> 00:38:51,200
that gets created by the law
and gets imposed or wrapped

797
00:38:51,200 --> 00:38:56,310
around the delivery
of that drink.

798
00:38:56,310 --> 00:38:56,810
OK.

799
00:38:56,810 --> 00:38:57,750
That's their first point.

800
00:38:57,750 --> 00:38:59,125
The second point
is to think more

801
00:38:59,125 --> 00:39:01,220
about what these
implied terms imply.

802
00:39:03,970 --> 00:39:10,150
Because if these implied terms
are implied by a legal system,

803
00:39:10,150 --> 00:39:12,700
then that means the
legal system has

804
00:39:12,700 --> 00:39:16,340
an interest in your contract.

805
00:39:16,340 --> 00:39:20,940
Legal system is
not undisinterested

806
00:39:20,940 --> 00:39:21,690
in your contract.

807
00:39:21,690 --> 00:39:23,370
So we should always
think that our contract

808
00:39:23,370 --> 00:39:24,630
is going to have two parties.

809
00:39:24,630 --> 00:39:28,080
We call them the promisor
and the promisee.

810
00:39:28,080 --> 00:39:29,910
I've told you
we've had contracts

811
00:39:29,910 --> 00:39:30,955
that can be performance.

812
00:39:30,955 --> 00:39:32,580
"The performancer"
isn't really a word.

813
00:39:32,580 --> 00:39:34,710
But let's just say
promisor and promisee.

814
00:39:34,710 --> 00:39:36,160
But there's always
a third party,

815
00:39:36,160 --> 00:39:39,060
which is the state, who is
in the middle of the contract

816
00:39:39,060 --> 00:39:44,430
in the sense that the state
will police many contracts

817
00:39:44,430 --> 00:39:48,130
and decide whether the state
likes the contract or not.

818
00:39:48,130 --> 00:39:51,060
And if the state doesn't
like the contract,

819
00:39:51,060 --> 00:39:53,700
then the state won't
enforce the contract.

820
00:39:53,700 --> 00:39:55,440
Or the state might
actually punish you

821
00:39:55,440 --> 00:39:56,800
because of the contract.

822
00:39:56,800 --> 00:40:01,380
So for example, the state cares
about the kinds of contracts.

823
00:40:01,380 --> 00:40:03,540
You can have contracts
for the sale of tables.

824
00:40:03,540 --> 00:40:05,800
You can't have contracts
for the sale of people.

825
00:40:05,800 --> 00:40:08,340
But you can sell dogs for
reasons I can't understand.

826
00:40:08,340 --> 00:40:11,310
But the point is we are
deciding which type of things

827
00:40:11,310 --> 00:40:14,610
can be sold and which kind
of things can't be sold.

828
00:40:14,610 --> 00:40:18,780
The state cares about the
effect of the contract.

829
00:40:18,780 --> 00:40:22,650
If the contract is to
render your corporation

830
00:40:22,650 --> 00:40:25,493
vulnerable to
bankruptcy, the state

831
00:40:25,493 --> 00:40:26,910
might have an
interest in deciding

832
00:40:26,910 --> 00:40:28,700
whether that contract
will be enforced

833
00:40:28,700 --> 00:40:30,450
or not or whether it'll
be allowed or not,

834
00:40:30,450 --> 00:40:34,140
whether it's permissible
under bankruptcy laws

835
00:40:34,140 --> 00:40:36,390
to engage in that contract
because of the risk

836
00:40:36,390 --> 00:40:37,610
that it's going to create.

837
00:40:37,610 --> 00:40:41,900
The state's going to care about
the terms of the contract.

838
00:40:41,900 --> 00:40:48,700
So if you're selling your labor
in a state with a minimum wage

839
00:40:48,700 --> 00:40:50,110
law, the state's going to care.

840
00:40:50,110 --> 00:40:54,100
Does the term that
specifies your wage equal

841
00:40:54,100 --> 00:40:56,680
or exceed the minimum wage?

842
00:40:56,680 --> 00:40:59,530
And states can obviously
care-- the most important thing

843
00:40:59,530 --> 00:41:01,300
for the state is whether
and how the state

844
00:41:01,300 --> 00:41:03,440
taxes the transaction
in the contract.

845
00:41:03,440 --> 00:41:07,463
So when does it tax
the transaction?

846
00:41:07,463 --> 00:41:09,380
What is the event that's
going to manifest it?

847
00:41:09,380 --> 00:41:11,360
And of course, the
contract can try

848
00:41:11,360 --> 00:41:15,148
to play with that sort of deal
with whether it will be taxed.

849
00:41:15,148 --> 00:41:17,690
And the state will care about
how the contract deals with it.

850
00:41:17,690 --> 00:41:18,190
OK.

851
00:41:18,190 --> 00:41:24,120
So the point is to fight
against the first real bias

852
00:41:24,120 --> 00:41:28,220
that especially,
let's say, tech people

853
00:41:28,220 --> 00:41:29,900
and maybe business
tech people bring

854
00:41:29,900 --> 00:41:32,563
to the idea of contract law,
which is that contracts are not

855
00:41:32,563 --> 00:41:33,230
about the state.

856
00:41:33,230 --> 00:41:35,480
They're about private parties.

857
00:41:35,480 --> 00:41:36,230
That's not true.

858
00:41:36,230 --> 00:41:39,290
They're about private
parties and the state.

859
00:41:39,290 --> 00:41:41,470
And the state is always
going to be there.

860
00:41:41,470 --> 00:41:44,810
Now, so if I say
to you, I'm trying

861
00:41:44,810 --> 00:41:46,010
to sell my parents' house.

862
00:41:46,010 --> 00:41:46,820
I can't sell it.

863
00:41:46,820 --> 00:41:48,290
So if I say to
you, I'll offer you

864
00:41:48,290 --> 00:41:52,460
$10,000 to anyone who will
burn down my parents' house.

865
00:41:52,460 --> 00:41:54,752
You're trying to
accept that contract.

866
00:41:54,752 --> 00:41:56,960
You're raising your hand,
so you're trying to accept.

867
00:41:56,960 --> 00:41:59,510
I want to say officially
it's a joke so nobody gets

868
00:41:59,510 --> 00:42:00,953
any uncertainty about this.

869
00:42:00,953 --> 00:42:02,870
Now, you're going to ask
a question, because I

870
00:42:02,870 --> 00:42:03,953
didn't want you to accept.

871
00:42:03,953 --> 00:42:06,350
Because if you accepted
my offer, then that's it.

872
00:42:06,350 --> 00:42:06,980
I'm stuck.

873
00:42:06,980 --> 00:42:09,230
So now it's clear it's
not really an offer,

874
00:42:09,230 --> 00:42:11,460
and I can ask you a question.

875
00:42:11,460 --> 00:42:14,517
AUDIENCE: So you said contracts
cannot be between two parties.

876
00:42:14,517 --> 00:42:16,600
They have to be between
two parties and the state.

877
00:42:16,600 --> 00:42:17,308
LARRY LESSIG: No.

878
00:42:17,308 --> 00:42:19,040
So let me say it more clearly.

879
00:42:19,040 --> 00:42:22,610
Contracts always have
the state present,

880
00:42:22,610 --> 00:42:24,198
so it's always
between two parties

881
00:42:24,198 --> 00:42:25,240
or three parties or four.

882
00:42:25,240 --> 00:42:26,900
So you and I enter
into a contract.

883
00:42:26,900 --> 00:42:29,870
Not this one, but another one.

884
00:42:29,870 --> 00:42:33,350
But the state is in
the room and deciding

885
00:42:33,350 --> 00:42:36,450
whether the state's going to
allow the contract to happen.

886
00:42:36,450 --> 00:42:38,450
So most contracts, the
state doesn't care about.

887
00:42:38,450 --> 00:42:40,370
But some contracts, the
state's going to say,

888
00:42:40,370 --> 00:42:43,600
hell no, we're not going
to allow it, like this one.

889
00:42:43,600 --> 00:42:46,347
If you didn't get that
notice it was a joke

890
00:42:46,347 --> 00:42:48,430
and you thought I was
serious and you went and you

891
00:42:48,430 --> 00:42:51,097
burned down my parents' house so
that we could get the insurance

892
00:42:51,097 --> 00:42:53,410
and not have to worry
about selling the house,

893
00:42:53,410 --> 00:42:56,590
and then you came to me, and
you said, OK, pay the $10,000.

894
00:42:56,590 --> 00:42:58,240
And I said, it was
an obvious joke.

895
00:42:58,240 --> 00:43:00,850
And you went to a court,
and you said, force Lessig

896
00:43:00,850 --> 00:43:02,800
to pay the $10,000.

897
00:43:02,800 --> 00:43:05,450
The court will say this
is an illegal contract.

898
00:43:05,450 --> 00:43:07,390
I'm not going to
enforce this contract.

899
00:43:07,390 --> 00:43:09,550
So my point is the state,
in a certain sense,

900
00:43:09,550 --> 00:43:12,830
is the censor of this contract.

901
00:43:12,830 --> 00:43:15,640
The state is in the room
when we make this contract.

902
00:43:15,640 --> 00:43:17,170
And the state's
judgment about it

903
00:43:17,170 --> 00:43:20,390
will decide whether
we enforce it or not.

904
00:43:20,390 --> 00:43:22,260
Is that clearer?

905
00:43:22,260 --> 00:43:24,802
AUDIENCE: So you're not saying
the state has to be there.

906
00:43:24,802 --> 00:43:27,010
LARRY LESSIG: I'm not saying
the state is technically

907
00:43:27,010 --> 00:43:28,600
signing the document.

908
00:43:28,600 --> 00:43:30,550
But I'm getting you
to recognize the fact

909
00:43:30,550 --> 00:43:35,130
that the contract is valuable
to the extent it's enforceable.

910
00:43:35,130 --> 00:43:38,870
And the state is the
essential agent in enforcement

911
00:43:38,870 --> 00:43:40,365
in the real world.

912
00:43:40,365 --> 00:43:42,740
AUDIENCE: So I guess that's
my question, because it seems

913
00:43:42,740 --> 00:43:46,510
like nowadays, we can have
contracts where the enforcement

914
00:43:46,510 --> 00:43:48,380
is not the state.

915
00:43:48,380 --> 00:43:50,190
I can put a contract
out there that says,

916
00:43:50,190 --> 00:43:51,860
hey, if you can
factor this number,

917
00:43:51,860 --> 00:43:54,664
give me the prime factors of
it, I'll give you 10 Ether.

918
00:43:54,664 --> 00:43:55,456
LARRY LESSIG: Yeah.

919
00:43:55,456 --> 00:43:57,950
And what I started with
when I was showing you

920
00:43:57,950 --> 00:44:01,100
this picture was to
say, actually, that's

921
00:44:01,100 --> 00:44:02,900
been true for a while, right?

922
00:44:02,900 --> 00:44:05,925
So in what sense
is the state here?

923
00:44:05,925 --> 00:44:08,050
Well, the state is here if
when you put the dime in

924
00:44:08,050 --> 00:44:10,400
and you get the drink out and
the drink doesn't have Dr.

925
00:44:10,400 --> 00:44:12,467
Pepper but it has
gasoline in it,

926
00:44:12,467 --> 00:44:14,300
the state would come
in and say, whoa, wait.

927
00:44:14,300 --> 00:44:17,090
You've breached the
implied term that said

928
00:44:17,090 --> 00:44:19,110
that this was safe to consume.

929
00:44:19,110 --> 00:44:21,980
So the state is in
this contract, too.

930
00:44:21,980 --> 00:44:24,680
And so what I'm trying to
lead to or get to the place

931
00:44:24,680 --> 00:44:27,770
where we say, is there ever a
place where the state is not

932
00:44:27,770 --> 00:44:33,950
there, which is what I think
your blockchain-like invocation

933
00:44:33,950 --> 00:44:34,760
is.

934
00:44:34,760 --> 00:44:38,140
So that's a question we're
going to get to in a second.

935
00:44:38,140 --> 00:44:39,140
Here's another contract.

936
00:44:42,380 --> 00:44:47,570
We'll offer you $50,000 to
lobby Congress to pass HR102.

937
00:44:47,570 --> 00:44:50,540
This contract looks
pretty normal to us.

938
00:44:50,540 --> 00:44:53,870
You can imagine lobbyists
have a contract like this

939
00:44:53,870 --> 00:44:55,440
all the time.

940
00:44:55,440 --> 00:44:57,380
In fact, in the middle
of the 19th century,

941
00:44:57,380 --> 00:44:59,270
the Supreme Court
ruled this contract

942
00:44:59,270 --> 00:45:01,130
was an illegal contract.

943
00:45:01,130 --> 00:45:04,590
You weren't allowed to hire
people to lobby Congress,

944
00:45:04,590 --> 00:45:05,090
right?

945
00:45:05,090 --> 00:45:07,490
So again, the
background norm of what

946
00:45:07,490 --> 00:45:11,030
is appropriate or not for the
contract controlled what type

947
00:45:11,030 --> 00:45:12,270
of contracting was allowed.

948
00:45:12,270 --> 00:45:12,770
OK.

949
00:45:12,770 --> 00:45:15,020
So that's about the
type of contract.

950
00:45:15,020 --> 00:45:17,160
But then think about the
terms of the contract.

951
00:45:17,160 --> 00:45:20,090
And so now, I want to
get you to recognize

952
00:45:20,090 --> 00:45:25,430
the way in which the terms
that the law wants a contract

953
00:45:25,430 --> 00:45:30,383
to have can be defeated
or not by technology.

954
00:45:30,383 --> 00:45:32,550
And the way I'm going to
get you to think about that

955
00:45:32,550 --> 00:45:35,490
is to think about not contract
law but something that

956
00:45:35,490 --> 00:45:37,530
will create the intuitions
I want you to have.

957
00:45:37,530 --> 00:45:38,880
Think about copyright law.

958
00:45:38,880 --> 00:45:40,440
Everybody knows
copyright law, right?

959
00:45:40,440 --> 00:45:43,720
I hope you know something
about copyright law.

960
00:45:43,720 --> 00:45:47,310
So copyright law is a basic
contract with the state.

961
00:45:47,310 --> 00:45:51,570
But the state says, if
you create something,

962
00:45:51,570 --> 00:45:54,360
then you're going to get
an exclusive right to it

963
00:45:54,360 --> 00:45:56,480
or exclusive rights,
a set of rights--

964
00:45:56,480 --> 00:45:59,520
an inclusive right to copy,
an exclusive right to sell it,

965
00:45:59,520 --> 00:46:03,330
exclusive right to make
derivative works of it--

966
00:46:03,330 --> 00:46:05,270
for a period of time--

967
00:46:05,270 --> 00:46:10,530
and in America right now,
that's your life plus 70 years--

968
00:46:10,530 --> 00:46:13,620
subject to fair
use, meaning the law

969
00:46:13,620 --> 00:46:17,900
says you can't control every
use of my copyrighted work.

970
00:46:17,900 --> 00:46:21,480
If you want to take a book
of mine and quote a section

971
00:46:21,480 --> 00:46:24,720
and write a review that says,
see how idiotic Lessig is?

972
00:46:24,720 --> 00:46:25,940
You can do that.

973
00:46:25,940 --> 00:46:28,020
And I can't wrap the
book in a contract that

974
00:46:28,020 --> 00:46:30,990
says you're not allowed to quote
it for purposes of criticizing.

975
00:46:30,990 --> 00:46:33,150
I'm just not allowed to do that.

976
00:46:33,150 --> 00:46:36,450
So copyright law, imagine
this bundle of rights

977
00:46:36,450 --> 00:46:38,370
that it was offering.

978
00:46:38,370 --> 00:46:41,870
But now imagine something called
DRM, Digital Rights Management.

979
00:46:41,870 --> 00:46:42,370
All right.

980
00:46:42,370 --> 00:46:48,050
So DRM is a set of code that we
can wrap copyrighted material

981
00:46:48,050 --> 00:46:52,040
in in the process of making
it available to others

982
00:46:52,040 --> 00:46:56,183
across networks
and whatever else.

983
00:46:56,183 --> 00:46:57,600
It's pretty trivial
to see the way

984
00:46:57,600 --> 00:47:03,730
DRM could, in effect, destroy
the limitations on the term.

985
00:47:03,730 --> 00:47:06,550
You can wrap it,
encrypt it, make it

986
00:47:06,550 --> 00:47:08,860
so that my ability to
control it will be my ability

987
00:47:08,860 --> 00:47:13,370
to control it, conceivably, as
long as machines are running.

988
00:47:13,370 --> 00:47:15,640
If it's a
Microsoft-based system,

989
00:47:15,640 --> 00:47:17,360
four years later, it
won't be workable.

990
00:47:17,360 --> 00:47:20,680
But the point is, in
principle, it could be forever.

991
00:47:20,680 --> 00:47:22,630
Same thing with fair use.

992
00:47:22,630 --> 00:47:26,110
You can imagine it being
wrapped in a way that

993
00:47:26,110 --> 00:47:30,470
disables the capacity
to engage in fair use.

994
00:47:30,470 --> 00:47:36,090
So I make tons of
presentations all the time

995
00:47:36,090 --> 00:47:39,030
where I need to capture video.

996
00:47:39,030 --> 00:47:42,240
And I was astonished
to discover the latest

997
00:47:42,240 --> 00:47:46,850
round of this operating
system basically

998
00:47:46,850 --> 00:47:50,000
makes it incredibly
difficult to rip

999
00:47:50,000 --> 00:47:52,310
video for the purpose
of capturing even two

1000
00:47:52,310 --> 00:47:53,190
seconds of it.

1001
00:47:53,190 --> 00:47:56,850
So if you have a video program
that is capturing your screen,

1002
00:47:56,850 --> 00:47:59,540
this operating system
will now disable it

1003
00:47:59,540 --> 00:48:01,790
for any period of
time that you're

1004
00:48:01,790 --> 00:48:03,110
trying to capture the screen.

1005
00:48:03,110 --> 00:48:05,660
They've built the technology
so that, now, there

1006
00:48:05,660 --> 00:48:08,570
is no capacity,
technical capacity,

1007
00:48:08,570 --> 00:48:10,880
to engage in fair
use of copyrighted

1008
00:48:10,880 --> 00:48:13,880
material on the Apple platform.

1009
00:48:13,880 --> 00:48:15,680
Now, you might say,
why should they

1010
00:48:15,680 --> 00:48:19,610
be allowed to do that
if the law intended

1011
00:48:19,610 --> 00:48:24,567
that the contract gives you
the free use of fair use?

1012
00:48:24,567 --> 00:48:26,150
Why should they be
allowed to do that?

1013
00:48:26,150 --> 00:48:29,540
But that's the battle that's
been going on about DRM

1014
00:48:29,540 --> 00:48:32,750
ever since the beginning of
the beginning of copyrighted

1015
00:48:32,750 --> 00:48:34,100
work on the internet.

1016
00:48:34,100 --> 00:48:35,630
So my point is to
get you to realize

1017
00:48:35,630 --> 00:48:42,622
the way the code becomes part
of the law of this contract.

1018
00:48:42,622 --> 00:48:44,330
And you have to ask
yourself the question

1019
00:48:44,330 --> 00:48:49,340
whether that law is respecting
the law of the sovereign,

1020
00:48:49,340 --> 00:48:50,700
the law of the jurisdiction.

1021
00:48:50,700 --> 00:48:54,170
So when Apple and
Disney get together

1022
00:48:54,170 --> 00:48:56,420
and they sell this
technology and they

1023
00:48:56,420 --> 00:48:59,840
sell movies across
the iTunes platform

1024
00:48:59,840 --> 00:49:02,710
that make it
practically impossible--

1025
00:49:02,710 --> 00:49:03,860
there's not even code.

1026
00:49:03,860 --> 00:49:05,720
I mean, if you go out
there and you look,

1027
00:49:05,720 --> 00:49:08,095
you can find everybody who
says, yeah, with this version,

1028
00:49:08,095 --> 00:49:10,410
nobody's yet found
a way to break it.

1029
00:49:10,410 --> 00:49:12,650
So right now, we're in a
world where, right now, it's

1030
00:49:12,650 --> 00:49:15,260
not possible for me to
capture three seconds to put

1031
00:49:15,260 --> 00:49:17,840
into a presentation.

1032
00:49:17,840 --> 00:49:20,420
And the issue that
raises is, to what extent

1033
00:49:20,420 --> 00:49:23,450
is that consistent
with copyright law

1034
00:49:23,450 --> 00:49:25,370
if the technology
now disables you

1035
00:49:25,370 --> 00:49:27,400
from doing exactly what
copyright law wants?

1036
00:49:27,400 --> 00:49:30,440
So this is the trade-off
to think about,

1037
00:49:30,440 --> 00:49:35,180
the relationship between the
policy that the law wants

1038
00:49:35,180 --> 00:49:38,130
and the policy of
the technology.

1039
00:49:38,130 --> 00:49:40,810
And there's no reason to believe
the policy of the technology

1040
00:49:40,810 --> 00:49:44,575
will always be
consistent with the law.

1041
00:49:44,575 --> 00:49:46,700
And to the extent it's not
consistent with the law,

1042
00:49:46,700 --> 00:49:47,660
it challenges the law.

1043
00:49:47,660 --> 00:49:52,940
It says to the law, OK,
come out and get me.

1044
00:49:52,940 --> 00:49:54,920
Now, that obviously
is implicated

1045
00:49:54,920 --> 00:49:58,420
in the context of
smart contracts,

1046
00:49:58,420 --> 00:50:00,550
because the biggest fear
about smart contracts

1047
00:50:00,550 --> 00:50:05,230
is that they enable a kind
of transaction that can hide

1048
00:50:05,230 --> 00:50:07,910
from the policy of the law.

1049
00:50:07,910 --> 00:50:12,070
So the question for the law is,
what will you do to step in?

1050
00:50:12,070 --> 00:50:14,390
And this is, I think, where
your question was going.

1051
00:50:14,390 --> 00:50:17,970
How does the state appear in
the middle of that contract?

1052
00:50:17,970 --> 00:50:21,760
But let's take one more
step before we get to that.

1053
00:50:21,760 --> 00:50:22,260
OK.

1054
00:50:22,260 --> 00:50:23,450
That's the second point.

1055
00:50:23,450 --> 00:50:26,240
Here's the third point.

1056
00:50:26,240 --> 00:50:29,330
There's often this intuition
that, especially technologists,

1057
00:50:29,330 --> 00:50:31,720
people have about what the
particular aim of contract law

1058
00:50:31,720 --> 00:50:31,910
is.

1059
00:50:31,910 --> 00:50:33,410
So in the very
beginning of me writing

1060
00:50:33,410 --> 00:50:35,410
about the law of cyberspace,
somebody from MIT--

1061
00:50:35,410 --> 00:50:37,460
I won't name any
names-- it wasn't Andy--

1062
00:50:37,460 --> 00:50:40,140
showed up at the Harvard
Law School and said,

1063
00:50:40,140 --> 00:50:42,328
I've solved the problem
of contract law.

1064
00:50:42,328 --> 00:50:43,370
I said, what do you mean?

1065
00:50:43,370 --> 00:50:46,580
He said, I have a system
that will eliminate risks

1066
00:50:46,580 --> 00:50:48,562
in contracts.

1067
00:50:48,562 --> 00:50:50,770
And he was really disappointed
when I said, you know,

1068
00:50:50,770 --> 00:50:54,380
the objective of contract law
is not to eliminate risks.

1069
00:50:54,380 --> 00:50:57,010
The objective of contract
law is just to allocate risk,

1070
00:50:57,010 --> 00:50:59,722
to figure out who has
the risk, so each of us

1071
00:50:59,722 --> 00:51:01,680
can figure out what to
do in light of the risk.

1072
00:51:01,680 --> 00:51:05,540
So if I say I will buy
10,000 bushels of corn

1073
00:51:05,540 --> 00:51:07,520
from you at $350 a bushel--

1074
00:51:07,520 --> 00:51:09,680
turns out that is the
price of corn right now.

1075
00:51:09,680 --> 00:51:11,430
He always looks up
things on the internet.

1076
00:51:11,430 --> 00:51:13,000
I figure I should look it
up before I present it.

1077
00:51:13,000 --> 00:51:15,630
So $350 is the price of a bushel
of corn right now delivered

1078
00:51:15,630 --> 00:51:18,120
September 1.

1079
00:51:18,120 --> 00:51:20,610
What that's basically
doing is it's allocating

1080
00:51:20,610 --> 00:51:22,510
the risk of the price of corn.

1081
00:51:22,510 --> 00:51:24,390
So if you're a
farmer and you don't

1082
00:51:24,390 --> 00:51:28,080
want to face the risk the price
of corn is going to fall to $3,

1083
00:51:28,080 --> 00:51:30,480
you enter into this contract.

1084
00:51:30,480 --> 00:51:33,600
If you're a buyer, you
accept this contract.

1085
00:51:33,600 --> 00:51:35,670
But if the price of
corn falls to $3,

1086
00:51:35,670 --> 00:51:37,210
well, you're out of luck.

1087
00:51:37,210 --> 00:51:42,960
So the point is it
relocates the price change,

1088
00:51:42,960 --> 00:51:44,880
reallocates the risk
of price change,

1089
00:51:44,880 --> 00:51:47,500
also reallocates the
risk of delivery.

1090
00:51:47,500 --> 00:51:51,210
So if I've got all this corn, I
can place it in a certain place

1091
00:51:51,210 --> 00:51:55,260
to shift the risk of delivery
so that, once again, I

1092
00:51:55,260 --> 00:51:58,350
use the contract to
shift my risk of storage

1093
00:51:58,350 --> 00:52:00,810
or my risk of delivery
and you on the other side.

1094
00:52:00,810 --> 00:52:03,570
So the objective of
contract law, number one,

1095
00:52:03,570 --> 00:52:06,090
is risk allocation.

1096
00:52:06,090 --> 00:52:06,590
Number one.

1097
00:52:06,590 --> 00:52:12,755
But allocation matters only if--

1098
00:52:12,755 --> 00:52:14,600
trying to say only if for me--

1099
00:52:14,600 --> 00:52:17,120
only if there is a
system to process

1100
00:52:17,120 --> 00:52:23,470
the breach of a contract, only
if you've got a technology,

1101
00:52:23,470 --> 00:52:25,243
let's say, for
processing the breach.

1102
00:52:25,243 --> 00:52:26,660
So if I have a
contract with you--

1103
00:52:26,660 --> 00:52:30,480
10,000 bushels of corn,
$3.50, on September 1--

1104
00:52:30,480 --> 00:52:35,040
and you don't deliver,
that's a breach.

1105
00:52:35,040 --> 00:52:38,400
My risk has only
been reallocated

1106
00:52:38,400 --> 00:52:40,800
if there's some way
for me to enforce

1107
00:52:40,800 --> 00:52:42,330
the contract at that point.

1108
00:52:42,330 --> 00:52:44,370
So it requires the system.

1109
00:52:44,370 --> 00:52:46,290
And the system we
ordinarily take for granted

1110
00:52:46,290 --> 00:52:48,680
is a legal system.

1111
00:52:48,680 --> 00:52:50,060
And I say we take
it for granted,

1112
00:52:50,060 --> 00:52:52,700
meaning some people can
take it for granted,

1113
00:52:52,700 --> 00:52:56,710
like people in well-developed
legal contexts.

1114
00:52:56,710 --> 00:52:59,250
But that's also to emphasized
other people can't take it

1115
00:52:59,250 --> 00:53:00,310
for granted--

1116
00:53:00,310 --> 00:53:03,580
so for example, people
in developing worlds.

1117
00:53:03,580 --> 00:53:05,620
They can't take the
legal system for granted.

1118
00:53:05,620 --> 00:53:07,790
They can't assume that if
the contract is breached,

1119
00:53:07,790 --> 00:53:09,873
they go into a court and
say, enforce the contract

1120
00:53:09,873 --> 00:53:11,380
or give me damages.

1121
00:53:11,380 --> 00:53:13,240
Or if you've got a
contract between somebody

1122
00:53:13,240 --> 00:53:15,982
in Rwanda and
somebody in Alaska,

1123
00:53:15,982 --> 00:53:17,440
then there, too,
there's a question

1124
00:53:17,440 --> 00:53:20,210
of whether we've got sufficient
infrastructure to develop.

1125
00:53:20,210 --> 00:53:24,010
So the point I want to
emphasize at the end

1126
00:53:24,010 --> 00:53:27,010
of this little
intervention is to suggest

1127
00:53:27,010 --> 00:53:30,730
that this fact that it
takes for granted the system

1128
00:53:30,730 --> 00:53:35,790
of enforcement shows the real
potential benefit of this class

1129
00:53:35,790 --> 00:53:39,220
of contracting devices.

1130
00:53:39,220 --> 00:53:41,490
Because if we think about
this as a representation

1131
00:53:41,490 --> 00:53:44,100
of a well-developed
legal system and we

1132
00:53:44,100 --> 00:53:47,760
think of first-world
countries as having those

1133
00:53:47,760 --> 00:53:50,550
and we compare third-world
or developing world

1134
00:53:50,550 --> 00:53:53,490
countries, which don't have
these well-developed legal

1135
00:53:53,490 --> 00:53:58,335
systems, you might say that
first-world countries are

1136
00:53:58,335 --> 00:54:00,210
better off relative to
third-world countries,

1137
00:54:00,210 --> 00:54:01,960
because they have the
legal infrastructure

1138
00:54:01,960 --> 00:54:04,003
to enable this risk
allocation that will enable

1139
00:54:04,003 --> 00:54:06,420
all sorts of market transactions
to happen which otherwise

1140
00:54:06,420 --> 00:54:08,170
wouldn't happen.

1141
00:54:08,170 --> 00:54:12,430
But if we imagine
infrastructure like

1142
00:54:12,430 --> 00:54:14,710
blockchain-like
infrastructures or Ethereum

1143
00:54:14,710 --> 00:54:18,190
infrastructures
to enter the mix,

1144
00:54:18,190 --> 00:54:20,530
providing a technical
infrastructure that

1145
00:54:20,530 --> 00:54:24,910
doesn't require judges and
lawyers but just requires code,

1146
00:54:24,910 --> 00:54:30,910
then that can substitute for
the legal infrastructure,

1147
00:54:30,910 --> 00:54:33,760
provide the infrastructure
the contract needs

1148
00:54:33,760 --> 00:54:36,520
at a much lower cost,
and thereby enable

1149
00:54:36,520 --> 00:54:41,380
people to contract who otherwise
wouldn't rationally contract,

1150
00:54:41,380 --> 00:54:45,730
because they could never
count on the infrastructure

1151
00:54:45,730 --> 00:54:49,730
of the legal system to
deliver what they need.

1152
00:54:49,730 --> 00:54:50,230
OK.

1153
00:54:50,230 --> 00:54:52,510
So the point is this
picture is trying

1154
00:54:52,510 --> 00:54:57,430
to say that one key value
of these technologies

1155
00:54:57,430 --> 00:55:00,370
is that they will be a
substitute for a failed

1156
00:55:00,370 --> 00:55:03,460
legal system or a legal system
that's not yet developed.

1157
00:55:03,460 --> 00:55:05,890
They will provide
the infrastructure

1158
00:55:05,890 --> 00:55:09,350
of a legal system or
the legal systems.

1159
00:55:09,350 --> 00:55:11,110
AUDIENCE: How can they
really do anything?

1160
00:55:11,110 --> 00:55:13,540
Because you can't make
that contract like,

1161
00:55:13,540 --> 00:55:19,108
I will pay you $3.50 per bushel
on September 1 in Ethereum,

1162
00:55:19,108 --> 00:55:21,400
because you're making a
statement about the real world.

1163
00:55:21,400 --> 00:55:25,180
The Ethereum blockchain
cannot verify that I gave you

1164
00:55:25,180 --> 00:55:25,680
the bushels.

1165
00:55:25,680 --> 00:55:26,388
LARRY LESSIG: OK.

1166
00:55:26,388 --> 00:55:28,850
So I've got two slides to get
to your example [INAUDIBLE]..

1167
00:55:28,850 --> 00:55:29,350
OK.

1168
00:55:29,350 --> 00:55:31,683
AUDIENCE: There's ways to do
it, but they're very risky,

1169
00:55:31,683 --> 00:55:32,500
I should say.

1170
00:55:32,500 --> 00:55:33,333
LARRY LESSIG: Right.

1171
00:55:33,333 --> 00:55:35,320
But let's bracket it for just--

1172
00:55:35,320 --> 00:55:37,740
I'm not sure if it's
two, but it's n slides.

1173
00:55:37,740 --> 00:55:40,510
But n is less than
100, I promise.

1174
00:55:40,510 --> 00:55:43,240
So just hold one second, OK?

1175
00:55:43,240 --> 00:55:45,000
So if you say that
one function here

1176
00:55:45,000 --> 00:55:49,240
is to substitute for
a failed legal system,

1177
00:55:49,240 --> 00:55:55,400
that suggests that the
other key opportunity here

1178
00:55:55,400 --> 00:55:59,810
is to enable contracts where the
transaction costs right now are

1179
00:55:59,810 --> 00:56:02,480
otherwise too high.

1180
00:56:02,480 --> 00:56:05,530
So this is a key opportunity
to think of places where

1181
00:56:05,530 --> 00:56:09,400
if you can lower the transaction
costs of the contract,

1182
00:56:09,400 --> 00:56:11,343
you will enable a
contract which otherwise

1183
00:56:11,343 --> 00:56:12,260
can't exist right now.

1184
00:56:12,260 --> 00:56:18,040
So I'm on a scientific board
for an insurance company.

1185
00:56:18,040 --> 00:56:20,830
So I was meeting the
president last week.

1186
00:56:20,830 --> 00:56:24,057
And he was all excited
about a new product which

1187
00:56:24,057 --> 00:56:26,140
they were investing in,
which was going to provide

1188
00:56:26,140 --> 00:56:28,390
flight delay insurance.

1189
00:56:28,390 --> 00:56:32,770
And this product was going to
be completely blockchain driven.

1190
00:56:32,770 --> 00:56:35,410
And this is what I'm trying
to get to your point.

1191
00:56:35,410 --> 00:56:40,420
It would trigger payments in a
completely automatic way based

1192
00:56:40,420 --> 00:56:42,220
on the information
that's being reported

1193
00:56:42,220 --> 00:56:44,650
from the n different sites
that are reporting information

1194
00:56:44,650 --> 00:56:45,233
about flights.

1195
00:56:45,233 --> 00:56:47,020
So I buy the contract.

1196
00:56:47,020 --> 00:56:49,330
It says, if my flight is
delayed more than an hour,

1197
00:56:49,330 --> 00:56:51,250
I get paid $200.

1198
00:56:51,250 --> 00:56:54,910
And automatically,
as he put it, it's

1199
00:56:54,910 --> 00:56:56,980
a no-touch product,
by which he meant

1200
00:56:56,980 --> 00:56:59,740
no human needs to touch this
product for this contract

1201
00:56:59,740 --> 00:57:01,185
to be enabled.

1202
00:57:01,185 --> 00:57:02,560
It's a kind of
product that never

1203
00:57:02,560 --> 00:57:05,530
would have been available
before, because the transaction

1204
00:57:05,530 --> 00:57:07,742
costs of engaging it
were way too high.

1205
00:57:07,742 --> 00:57:09,700
But now that we've lowered
the transaction cost

1206
00:57:09,700 --> 00:57:11,800
and have an infrastructure
of trust, which

1207
00:57:11,800 --> 00:57:14,590
is what the blockchain
is providing here,

1208
00:57:14,590 --> 00:57:17,650
there's a huge
market that now is

1209
00:57:17,650 --> 00:57:20,350
available for a contract
that otherwise just

1210
00:57:20,350 --> 00:57:21,580
wasn't there before.

1211
00:57:21,580 --> 00:57:24,190
Now, that market depends on
making a couple assumptions.

1212
00:57:24,190 --> 00:57:27,552
You're not saying that
there's 100% certainty

1213
00:57:27,552 --> 00:57:29,260
that everything in
that market is working

1214
00:57:29,260 --> 00:57:31,130
the way it's supposed to work.

1215
00:57:31,130 --> 00:57:34,370
But the point is you
don't need 100% certainty

1216
00:57:34,370 --> 00:57:37,170
for the vast majority of
these kinds of contracts.

1217
00:57:37,170 --> 00:57:39,470
If it gets it wrong
every once in a while,

1218
00:57:39,470 --> 00:57:41,390
that's good enough
for government work

1219
00:57:41,390 --> 00:57:43,310
or good enough
for airplane work.

1220
00:57:43,310 --> 00:57:46,520
And so that's going to
be sufficient to enable

1221
00:57:46,520 --> 00:57:50,270
the market, recognizing that
even in the legal system

1222
00:57:50,270 --> 00:57:51,390
market, guess what?

1223
00:57:51,390 --> 00:57:54,170
It doesn't always
get it right either.

1224
00:57:54,170 --> 00:57:56,860
But it doesn't always enforce
the contract well either.

1225
00:57:56,860 --> 00:57:59,030
In fact, the costs
are much, much higher

1226
00:57:59,030 --> 00:58:00,650
to fail in the legal system.

1227
00:58:00,650 --> 00:58:01,340
OK.

1228
00:58:01,340 --> 00:58:03,440
So two kinds of
transaction costs.

1229
00:58:03,440 --> 00:58:06,950
One system transaction
cost suggests

1230
00:58:06,950 --> 00:58:10,310
that where you have less
developed legal systems,

1231
00:58:10,310 --> 00:58:12,770
the blockchain is going to
provide a real opportunity,

1232
00:58:12,770 --> 00:58:15,260
because you're going to
lower the transaction

1233
00:58:15,260 --> 00:58:16,730
costs for those systems.

1234
00:58:16,730 --> 00:58:19,390
And the other is just
contract transaction costs.

1235
00:58:19,390 --> 00:58:22,280
There'll be a huge
explosion of markets where,

1236
00:58:22,280 --> 00:58:26,060
right now, the contract
transaction costs are so high,

1237
00:58:26,060 --> 00:58:28,183
and we can lower them
and thereby enable

1238
00:58:28,183 --> 00:58:29,600
a certain kind of
transaction that

1239
00:58:29,600 --> 00:58:30,740
otherwise wouldn't be there.

1240
00:58:30,740 --> 00:58:31,240
OK.

1241
00:58:31,240 --> 00:58:32,750
Final point I want to make.

1242
00:58:32,750 --> 00:58:36,250
The other thing technologists
always love to say

1243
00:58:36,250 --> 00:58:37,910
is the great thing
we will do when

1244
00:58:37,910 --> 00:58:42,100
we do contract
technology is they'll

1245
00:58:42,100 --> 00:58:44,860
solve the clarity problem.

1246
00:58:44,860 --> 00:58:47,097
Every term will be
perfectly clear.

1247
00:58:47,097 --> 00:58:48,680
Because you write
those contracts out,

1248
00:58:48,680 --> 00:58:51,070
and there's all sorts of
ambiguity and vagueness,

1249
00:58:51,070 --> 00:58:52,060
and it's a total mess.

1250
00:58:52,060 --> 00:58:55,500
But when we've coded
all of our contracts,

1251
00:58:55,500 --> 00:58:58,060
our smart contracts,
then we'll have

1252
00:58:58,060 --> 00:59:00,490
perfect certainty for
every single outcome,

1253
00:59:00,490 --> 00:59:02,987
and that will be great.

1254
00:59:02,987 --> 00:59:04,570
And the point I want
to suggest to you

1255
00:59:04,570 --> 00:59:10,800
is that, often, obscurity
is a real value.

1256
00:59:10,800 --> 00:59:14,700
Obscurity is what you want,
because here's a way to see it.

1257
00:59:14,700 --> 00:59:16,210
So imagine this is
a decision tree.

1258
00:59:16,210 --> 00:59:17,430
And it's really
small, because I just

1259
00:59:17,430 --> 00:59:18,420
stole it from the internet.

1260
00:59:18,420 --> 00:59:19,920
And it doesn't have anything
to do really with what

1261
00:59:19,920 --> 00:59:20,878
I'm talking about here.

1262
00:59:20,878 --> 00:59:23,760
But imagine a really
complicated decision tree

1263
00:59:23,760 --> 00:59:26,850
like this, which is to represent
all the possible things that

1264
00:59:26,850 --> 00:59:28,950
could happen with our deal.

1265
00:59:28,950 --> 00:59:30,240
So I want to buy the house.

1266
00:59:30,240 --> 00:59:33,060
But what happens if the
house gets hit by a meteor?

1267
00:59:33,060 --> 00:59:35,040
Or what happens if
you go bankrupt?

1268
00:59:35,040 --> 00:59:37,980
All of these possible outcomes.

1269
00:59:37,980 --> 00:59:40,950
And in principle, we could
say we should be negotiating

1270
00:59:40,950 --> 00:59:42,217
every one of these blue dots.

1271
00:59:42,217 --> 00:59:44,300
We should be saying, what
happens if this happens?

1272
00:59:44,300 --> 00:59:46,230
OK, then this is what you
get, and this is what I get.

1273
00:59:46,230 --> 00:59:47,130
What happens if that happens?

1274
00:59:47,130 --> 00:59:47,960
What you get, what I get.

1275
00:59:47,960 --> 00:59:48,460
OK.

1276
00:59:48,460 --> 00:59:50,730
So imagine this
blue dot here has

1277
00:59:50,730 --> 00:59:55,260
a 0.002% chance of happening.

1278
00:59:55,260 --> 00:59:58,870
So you know the chance of that
happening is practically zero.

1279
00:59:58,870 --> 01:00:02,620
And so what's the
reasonable amount of time

1280
01:00:02,620 --> 01:00:06,420
you should spend
negotiating that term?

1281
01:00:06,420 --> 01:00:08,320
Zero, right?

1282
01:00:08,320 --> 01:00:10,840
Especially because if
this term is something

1283
01:00:10,840 --> 01:00:14,775
you know he's really
worked up about

1284
01:00:14,775 --> 01:00:17,150
and you could never come to
an agreement about that term,

1285
01:00:17,150 --> 01:00:19,660
it would be ridiculous that
the whole contract fall apart

1286
01:00:19,660 --> 01:00:23,440
because of this
0.002% possibility.

1287
01:00:23,440 --> 01:00:26,650
So what contracts
do all the time

1288
01:00:26,650 --> 01:00:28,990
is they create
these fuzzy or vague

1289
01:00:28,990 --> 01:00:32,950
or ambiguous places as a gamble.

1290
01:00:32,950 --> 01:00:35,210
It's like, OK, we'll go forward.

1291
01:00:35,210 --> 01:00:39,730
We're going to just gamble
that this 0.002% outcome won't

1292
01:00:39,730 --> 01:00:41,740
happen.

1293
01:00:41,740 --> 01:00:44,950
Most of the time, in fact,
you can work out how often

1294
01:00:44,950 --> 01:00:46,475
it doesn't happen, right?

1295
01:00:46,475 --> 01:00:48,350
So most of the time,
that would be just fine.

1296
01:00:48,350 --> 01:00:52,360
And if it turns out to
happen, then what we'll do

1297
01:00:52,360 --> 01:00:55,090
is ask somebody--

1298
01:00:55,090 --> 01:00:59,420
namely, a judge-- to
figure that term out.

1299
01:00:59,420 --> 01:01:02,050
We'll say, what should that
term be interpreted as?

1300
01:01:02,050 --> 01:01:04,240
It was ambiguous, so what
should the answer be?

1301
01:01:04,240 --> 01:01:05,490
And the judge will look at it.

1302
01:01:05,490 --> 01:01:08,730
And judge will say, well, it's
fair that you win or he wins.

1303
01:01:08,730 --> 01:01:10,460
So the point is
there are many times

1304
01:01:10,460 --> 01:01:12,002
when you ask the
question, should you

1305
01:01:12,002 --> 01:01:14,750
negotiate a term ex ante?

1306
01:01:14,750 --> 01:01:17,760
And the answer to
that question is no.

1307
01:01:17,760 --> 01:01:19,640
And in those
contexts, an ambiguity

1308
01:01:19,640 --> 01:01:24,230
is a way to negotiate it ex
post, meaning after the fact,

1309
01:01:24,230 --> 01:01:27,790
ex post, by using the court.

1310
01:01:27,790 --> 01:01:31,710
So as we think about these
digital smart contracts,

1311
01:01:31,710 --> 01:01:35,120
this is why smart contracts are
referred to as dumb contracts.

1312
01:01:35,120 --> 01:01:37,100
They're smart contracts
for the equivalent

1313
01:01:37,100 --> 01:01:39,110
of dropping a dime
in and getting a Dr.

1314
01:01:39,110 --> 01:01:43,250
Pepper, the sorts of things
where the possibilities are

1315
01:01:43,250 --> 01:01:48,230
really small and narrow, or,
like my CEO was really excited,

1316
01:01:48,230 --> 01:01:49,670
is your plane late?

1317
01:01:49,670 --> 01:01:51,230
Yes, I get $50.

1318
01:01:51,230 --> 01:01:52,610
No, I don't.

1319
01:01:52,610 --> 01:01:56,480
Things where it's
relatively clear.

1320
01:01:56,480 --> 01:01:59,360
But there's a really
important conceptual question

1321
01:01:59,360 --> 01:02:00,860
about what happens
when we're trying

1322
01:02:00,860 --> 01:02:05,510
to use them for the kinds of
contracts where what we want

1323
01:02:05,510 --> 01:02:07,190
is ambiguity.

1324
01:02:07,190 --> 01:02:14,000
Because if the very terms of
deployment of the platform

1325
01:02:14,000 --> 01:02:20,573
demand specifying all of the
outcomes in this tree, then

1326
01:02:20,573 --> 01:02:22,490
your basic thing is a
whole bunch of contracts

1327
01:02:22,490 --> 01:02:24,980
you're just not going to be
able to have in this space,

1328
01:02:24,980 --> 01:02:29,570
because the cost of specifying
that tree will be higher

1329
01:02:29,570 --> 01:02:31,500
than the benefit
of those contracts.

1330
01:02:31,500 --> 01:02:34,670
So it's a weird sense in
which, on the one hand,

1331
01:02:34,670 --> 01:02:37,700
I've said to you that
this technology can

1332
01:02:37,700 --> 01:02:41,610
lower the transaction
costs of contracts.

1333
01:02:41,610 --> 01:02:43,550
But for this kind
of contract, it

1334
01:02:43,550 --> 01:02:47,710
can actually increase the
transaction costs of contract.

1335
01:02:47,710 --> 01:02:49,700
Because for this
kind of contract,

1336
01:02:49,700 --> 01:02:54,010
if it reveals the ambiguities
we need to negotiate about

1337
01:02:54,010 --> 01:02:55,990
and those negotiating
ambiguities are just

1338
01:02:55,990 --> 01:02:57,940
too costly for us
to negotiate, then

1339
01:02:57,940 --> 01:03:00,160
it's going to block us
from having that contract.

1340
01:03:00,160 --> 01:03:03,010
It forces us to see the
0.002%, and you and I

1341
01:03:03,010 --> 01:03:04,660
have to come to terms on it.

1342
01:03:04,660 --> 01:03:09,480
And we don't yet have a
good way to fake ambiguity.

1343
01:03:09,480 --> 01:03:11,880
I mean, the legal
system fakes ambiguity,

1344
01:03:11,880 --> 01:03:14,370
because we just say, oh, yeah,
yeah, we didn't see that.

1345
01:03:14,370 --> 01:03:15,590
They saw it.

1346
01:03:15,590 --> 01:03:17,640
He knew it when they
wrote the contract.

1347
01:03:17,640 --> 01:03:19,690
But they don't have
to admit they knew it.

1348
01:03:19,690 --> 01:03:25,110
But the code has to admit it
knows there's an ambiguity.

1349
01:03:25,110 --> 01:03:27,400
And that's a hard thing to
self consciously prevent.

1350
01:03:27,400 --> 01:03:27,900
OK.

1351
01:03:27,900 --> 01:03:29,330
So I've given you four ideas.

1352
01:03:29,330 --> 01:03:31,275
Happy to take
questions or abuse.

1353
01:03:31,275 --> 01:03:33,400
Usually, I get abuse from
business school students,

1354
01:03:33,400 --> 01:03:34,770
but OK.

1355
01:03:34,770 --> 01:03:40,530
AUDIENCE: So what if you can
code into a contract the easy

1356
01:03:40,530 --> 01:03:43,497
scenarios and then say else-if--

1357
01:03:43,497 --> 01:03:44,580
LARRY LESSIG: Go to court.

1358
01:03:44,580 --> 01:03:47,820
AUDIENCE: --go to court or
some multinational court

1359
01:03:47,820 --> 01:03:50,598
that is the Ethereum
court or whatever.

1360
01:03:50,598 --> 01:03:51,390
LARRY LESSIG: Yeah.

1361
01:03:51,390 --> 01:03:56,130
And even the Ethereum court will
have to have a court above it

1362
01:03:56,130 --> 01:03:58,760
if it wants to avoid government
saying, if you have anything

1363
01:03:58,760 --> 01:04:01,010
to do with Ethereum, we're
going to punish you, right?

1364
01:04:01,010 --> 01:04:03,093
So this is the sense of
which I wanted to suggest,

1365
01:04:03,093 --> 01:04:05,670
to go back to your
first question,

1366
01:04:05,670 --> 01:04:10,170
that if you think about it as
layers of an onion, peel layers

1367
01:04:10,170 --> 01:04:12,420
back enough, and you're
going to, in the end,

1368
01:04:12,420 --> 01:04:17,650
have to get to a state
in some sense, a state

1369
01:04:17,650 --> 01:04:19,860
as an enforceable mechanism.

1370
01:04:19,860 --> 01:04:22,500
And that enforceable
mechanism doesn't

1371
01:04:22,500 --> 01:04:24,030
have to be for every case.

1372
01:04:24,030 --> 01:04:26,190
Again, 99% of the time,
you drop the dime in,

1373
01:04:26,190 --> 01:04:27,690
you get the Dr. Pepper.

1374
01:04:27,690 --> 01:04:30,740
But there's a contingency where
you drop the dime in, and out

1375
01:04:30,740 --> 01:04:32,220
comes a bottle of gasoline.

1376
01:04:32,220 --> 01:04:34,053
And then you need to
go to somebody and say,

1377
01:04:34,053 --> 01:04:35,280
you breached this contract.

1378
01:04:35,280 --> 01:04:36,300
And that person--

1379
01:04:36,300 --> 01:04:39,390
AUDIENCE: But if it's a contract
between somebody in Alaska

1380
01:04:39,390 --> 01:04:41,190
and somebody in
Rwanda, you still

1381
01:04:41,190 --> 01:04:42,490
have the question of
which state you go to.

1382
01:04:42,490 --> 01:04:43,400
What's the jurisdiction?

1383
01:04:43,400 --> 01:04:44,942
So I feel like there
would have to be

1384
01:04:44,942 --> 01:04:48,078
some extranational extra-state.

1385
01:04:48,078 --> 01:04:48,870
LARRY LESSIG: Yeah.

1386
01:04:48,870 --> 01:04:51,090
So you will see in
these types of contracts

1387
01:04:51,090 --> 01:04:54,180
is there'll be a whole bunch
of boilerplate that specifies

1388
01:04:54,180 --> 01:04:56,940
choice of law, which turns out
to be a relatively easy thing

1389
01:04:56,940 --> 01:04:59,720
to do right now, choice
of jurisdictions.

1390
01:04:59,720 --> 01:05:01,310
For most countries,
it's pretty easy.

1391
01:05:01,310 --> 01:05:02,768
There are standard
ways to do that.

1392
01:05:02,768 --> 01:05:04,770
All of that will plug
in automatically.

1393
01:05:04,770 --> 01:05:07,710
But again, all of those
are outside of the code.

1394
01:05:07,710 --> 01:05:10,550
All of those are law.

1395
01:05:10,550 --> 01:05:11,940
AUDIENCE: Comment
and a question.

1396
01:05:11,940 --> 01:05:18,340
So I don't trust Vatalik
or other software engineers

1397
01:05:18,340 --> 01:05:22,020
that they have the legal
understanding and capacity

1398
01:05:22,020 --> 01:05:24,660
to write legally
binding contracts.

1399
01:05:24,660 --> 01:05:29,610
So I trust the government
or I trust the court.

1400
01:05:29,610 --> 01:05:31,500
I don't trust a
software engineer

1401
01:05:31,500 --> 01:05:36,850
to write a proper code to
protect me or my counterparty.

1402
01:05:36,850 --> 01:05:43,030
So what is your take on
those smart contracts?

1403
01:05:43,030 --> 01:05:45,210
PROFESSOR: Can I add to that?

1404
01:05:45,210 --> 01:05:48,630
Not only your take on
whether to trust but what

1405
01:05:48,630 --> 01:05:54,000
the courts in Europe and the US
right now do if somebody says,

1406
01:05:54,000 --> 01:05:56,728
I wrote the contract,
and here's the code.

1407
01:05:56,728 --> 01:05:58,770
LARRY LESSIG: Well, so
when you say-- let me just

1408
01:05:58,770 --> 01:06:01,930
me understand the
question better.

1409
01:06:01,930 --> 01:06:02,430
I'm sorry.

1410
01:06:02,430 --> 01:06:03,635
Your name is--

1411
01:06:03,635 --> 01:06:04,260
AUDIENCE: Hugo.

1412
01:06:04,260 --> 01:06:05,052
LARRY LESSIG: Hugo.

1413
01:06:05,052 --> 01:06:06,590
The way Hugo
described it, I think,

1414
01:06:06,590 --> 01:06:10,410
is nice, with a whole bunch of
conditional statements, which

1415
01:06:10,410 --> 01:06:11,280
is the code.

1416
01:06:11,280 --> 01:06:13,890
So the code says,
if the plane leaves

1417
01:06:13,890 --> 01:06:19,560
more than 30 minutes late, then
transfer $100 to his account.

1418
01:06:19,560 --> 01:06:21,840
So you don't trust
Vatalik's system

1419
01:06:21,840 --> 01:06:24,840
to be able to implement
that conditional?

1420
01:06:24,840 --> 01:06:25,800
AUDIENCE: Yep.

1421
01:06:25,800 --> 01:06:29,220
I don't understand-- I
mean, I don't know the code.

1422
01:06:29,220 --> 01:06:31,260
I don't know if it's breakable.

1423
01:06:31,260 --> 01:06:33,898
I don't know if Jean can
log in and change it.

1424
01:06:33,898 --> 01:06:35,190
LARRY LESSIG: I'm sure she can.

1425
01:06:35,190 --> 01:06:39,540
But the point is,
for most people,

1426
01:06:39,540 --> 01:06:41,550
this is true of everything.

1427
01:06:41,550 --> 01:06:43,110
When I say to you,
I'm a lawyer, I'm

1428
01:06:43,110 --> 01:06:47,130
going to write the contract so
you can sell your car to Andy,

1429
01:06:47,130 --> 01:06:48,817
you can have the same questions.

1430
01:06:48,817 --> 01:06:50,400
You can have the
question whether Andy

1431
01:06:50,400 --> 01:06:53,070
is going to have a way to
flip the price so instead

1432
01:06:53,070 --> 01:06:56,340
of $10,000, it's 15,000.

1433
01:06:56,340 --> 01:06:57,990
So this uncertainty
is everywhere.

1434
01:06:57,990 --> 01:07:00,510
It might be Vatalik
would say, actually,

1435
01:07:00,510 --> 01:07:04,500
there's a more robust
way to verify my code

1436
01:07:04,500 --> 01:07:06,772
than Lessig's contract code.

1437
01:07:06,772 --> 01:07:08,980
Because Lessig's contract
code is written in English,

1438
01:07:08,980 --> 01:07:10,230
and we have all sorts of--

1439
01:07:10,230 --> 01:07:12,390
but my code, you can
have independent people

1440
01:07:12,390 --> 01:07:13,230
who verify it.

1441
01:07:13,230 --> 01:07:18,090
So at the level of conditionals,
I'm pretty confident.

1442
01:07:18,090 --> 01:07:20,290
What I'm not
confident about is--

1443
01:07:20,290 --> 01:07:26,550
and I had the great pleasure
in December of 2015 spending

1444
01:07:26,550 --> 01:07:29,610
a weekend with them,
watching that group.

1445
01:07:29,610 --> 01:07:31,830
And it's this weird kind of--

1446
01:07:31,830 --> 01:07:35,040
he is like the messiah
figure, and there's

1447
01:07:35,040 --> 01:07:37,890
12 people sitting on the
floor around him listening

1448
01:07:37,890 --> 01:07:40,150
to his every bit of wisdom.

1449
01:07:40,150 --> 01:07:42,690
So it was inspirational
and scary at the same time,

1450
01:07:42,690 --> 01:07:44,400
because you see how
much money is resting

1451
01:07:44,400 --> 01:07:45,870
on this messiah's structure.

1452
01:07:45,870 --> 01:07:48,332
But I have a lot of
confidence in their ability

1453
01:07:48,332 --> 01:07:49,290
to do the conditionals.

1454
01:07:49,290 --> 01:07:51,390
But I'm not so
confident that they yet

1455
01:07:51,390 --> 01:07:55,200
thought through how it plugs
into the bigger legal story.

1456
01:07:55,200 --> 01:07:57,450
And that's part of what I
was trying to pitch to them,

1457
01:07:57,450 --> 01:07:58,440
that's what I'm trying--

1458
01:07:58,440 --> 01:07:59,857
to him, and that's
what I'm trying

1459
01:07:59,857 --> 01:08:02,926
to pitch to you, that it's
never going to be the case.

1460
01:08:02,926 --> 01:08:05,640
I was at this conference in
Australia where I met them.

1461
01:08:05,640 --> 01:08:08,140
And so many of the people there
were speaking the way people

1462
01:08:08,140 --> 01:08:09,723
spoke about the
internet 20 years ago.

1463
01:08:09,723 --> 01:08:12,250
It's like, oh, there's
the real world.

1464
01:08:12,250 --> 01:08:13,545
We have governments.

1465
01:08:13,545 --> 01:08:14,670
But we're going to create--

1466
01:08:14,670 --> 01:08:15,970
like John Perry Barlow speak.

1467
01:08:15,970 --> 01:08:18,106
We're going to create
this virtual world where

1468
01:08:18,106 --> 01:08:19,689
there is no governments,
where we just

1469
01:08:19,689 --> 01:08:21,040
live free of government.

1470
01:08:21,040 --> 01:08:26,200
And it's so fucking naive
about the way this stuff works,

1471
01:08:26,200 --> 01:08:29,470
because it's always embedded
in a real world which

1472
01:08:29,470 --> 01:08:30,687
has a real government.

1473
01:08:30,687 --> 01:08:32,229
And that real
government is not going

1474
01:08:32,229 --> 01:08:35,260
to look the other way if
you're doing lots of damage

1475
01:08:35,260 --> 01:08:36,620
or not paying your taxes.

1476
01:08:36,620 --> 01:08:38,500
So the question isn't whether.

1477
01:08:38,500 --> 01:08:40,936
The question is just
how and how much.

1478
01:08:40,936 --> 01:08:42,269
PROFESSOR: Can we go back there?

1479
01:08:42,269 --> 01:08:42,894
AUDIENCE: Yeah.

1480
01:08:42,894 --> 01:08:46,360
So I have a question about,
what kind of implications

1481
01:08:46,360 --> 01:08:50,470
do you think this technology
has to broaden access

1482
01:08:50,470 --> 01:08:53,649
to legal services in general,
like letting people know when

1483
01:08:53,649 --> 01:08:56,304
they're being taken advantage
or being able to provide more

1484
01:08:56,304 --> 01:08:58,179
standard versions of
really common contracts,

1485
01:08:58,179 --> 01:09:00,734
like employment contracts
or lease contacts

1486
01:09:00,734 --> 01:09:01,609
and things like that?

1487
01:09:01,609 --> 01:09:04,609
Do you think that this is going
to be a game changer for those?

1488
01:09:04,609 --> 01:09:06,340
LARRY LESSIG: Absolutely.

1489
01:09:06,340 --> 01:09:10,359
I mean, you've got to
muster the political will

1490
01:09:10,359 --> 01:09:13,770
to break the monopoly
lawyers have in this space.

1491
01:09:13,770 --> 01:09:16,359
And I'll tell you, we're going
to fight hard against it.

1492
01:09:16,359 --> 01:09:20,260
But there is no reason
for 90% of what lawyers

1493
01:09:20,260 --> 01:09:23,080
do to be done by lawyers.

1494
01:09:23,080 --> 01:09:26,050
Just like if you
buy a house and you

1495
01:09:26,050 --> 01:09:27,550
have to have a
lawyer who sits there

1496
01:09:27,550 --> 01:09:29,520
and basically signs
documents for you.

1497
01:09:29,520 --> 01:09:31,689
You're like, why?

1498
01:09:31,689 --> 01:09:33,228
And the answer is
because the law

1499
01:09:33,228 --> 01:09:35,770
has been written to require that
person to sit there and sign

1500
01:09:35,770 --> 01:09:36,279
documents.

1501
01:09:36,279 --> 01:09:36,700
And why?

1502
01:09:36,700 --> 01:09:38,742
Because the people who
get paid to sign documents

1503
01:09:38,742 --> 01:09:40,680
have corrupted the law.

1504
01:09:40,680 --> 01:09:42,370
So there's lots to clean up.

1505
01:09:42,370 --> 01:09:44,760
But if we can clean
it up, absolutely.

1506
01:09:44,760 --> 01:09:48,439
99% of this stuff
ought to be automatic.

1507
01:09:48,439 --> 01:09:51,770
And then let's get the lawyers
to focus on the hard corner

1508
01:09:51,770 --> 01:09:54,475
cases, not on everything
that should be automatic.

1509
01:09:54,475 --> 01:09:55,850
And that certainly
should be true

1510
01:09:55,850 --> 01:10:00,060
not just within a jurisdiction
but across jurisdiction.

1511
01:10:00,060 --> 01:10:03,640
AUDIENCE: Can I return to
your insurance examples?

1512
01:10:03,640 --> 01:10:07,270
I got the idea that you could
write a smart contract that

1513
01:10:07,270 --> 01:10:11,140
would pay you automatically
if the plane was late.

1514
01:10:11,140 --> 01:10:14,600
But then you said it
puts it on a blockchain.

1515
01:10:14,600 --> 01:10:17,380
Can you connect why he put it
on the blockchain as opposed

1516
01:10:17,380 --> 01:10:21,460
to just putting it on
his disc and letting

1517
01:10:21,460 --> 01:10:22,990
your credit card get credited?

1518
01:10:22,990 --> 01:10:25,870
I mean, where did the
blockchain enter into that?

1519
01:10:25,870 --> 01:10:27,790
And why did he need
a blockchain in order

1520
01:10:27,790 --> 01:10:31,233
to maintain that statement?

1521
01:10:31,233 --> 01:10:33,400
LARRY LESSIG: I think that
the thing that he thought

1522
01:10:33,400 --> 01:10:34,900
he was getting out
of the blockchain

1523
01:10:34,900 --> 01:10:39,130
is to increase the
verifiability of third parties

1524
01:10:39,130 --> 01:10:41,750
of the actual payments
according to the data.

1525
01:10:41,750 --> 01:10:44,380
So if I say to you, I'm
going to be paying you

1526
01:10:44,380 --> 01:10:47,230
$500 if your plane
is an hour late

1527
01:10:47,230 --> 01:10:51,610
and we have the data out there
about the planes being late,

1528
01:10:51,610 --> 01:10:54,040
if he did it inside of
his insurance company,

1529
01:10:54,040 --> 01:10:56,140
I wouldn't have any way
to know that he paid you

1530
01:10:56,140 --> 01:10:58,060
$500 for the plane being late.

1531
01:10:58,060 --> 01:11:00,880
But if it's in a blockchain,
there's at least conceptually

1532
01:11:00,880 --> 01:11:02,770
a way in which we
could be doing this--

1533
01:11:02,770 --> 01:11:04,300
not that I'm
reviewing it to you,

1534
01:11:04,300 --> 01:11:06,060
but we could be signaling--

1535
01:11:06,060 --> 01:11:07,310
AUDIENCE: That doesn't follow.

1536
01:11:07,310 --> 01:11:09,670
So that depends on who
maintains the blockchain.

1537
01:11:09,670 --> 01:11:12,220
And then at the same
time, it returns that.

1538
01:11:12,220 --> 01:11:14,290
Suddenly, everybody
knows the cash flow

1539
01:11:14,290 --> 01:11:16,480
of that insurance
company, which is likely

1540
01:11:16,480 --> 01:11:17,810
not something he wants to do.

1541
01:11:17,810 --> 01:11:20,310
LARRY LESSIG: Well, the cash
flow of the insurance company's

1542
01:11:20,310 --> 01:11:23,020
bigger than authenticating
the truth about certain types

1543
01:11:23,020 --> 01:11:23,660
of contracts.

1544
01:11:23,660 --> 01:11:24,160
AUDIENCE: Well, let
us know the cash flow

1545
01:11:24,160 --> 01:11:25,550
of that part of the business.

1546
01:11:25,550 --> 01:11:28,780
So I mean, he may not want that.

1547
01:11:28,780 --> 01:11:33,730
I may not want everybody to
know what planes I'm flying.

1548
01:11:33,730 --> 01:11:35,320
So you've implied
that there's now

1549
01:11:35,320 --> 01:11:38,610
a public blockchain that does
that to everyone to verify.

1550
01:11:38,610 --> 01:11:40,820
LARRY LESSIG: Yeah.

1551
01:11:40,820 --> 01:11:42,290
All I mean to argue--

1552
01:11:42,290 --> 01:11:43,700
and I didn't
interrogate him far.

1553
01:11:43,700 --> 01:11:45,700
I was just taking his
statement that he's doing.

1554
01:11:45,700 --> 01:11:47,825
So I'm trying to think,
why would he want to do it?

1555
01:11:47,825 --> 01:11:50,440
Because I had the same kind
of question you wanted to.

1556
01:11:50,440 --> 01:11:54,580
And it's just that if you are
trying to create credibility

1557
01:11:54,580 --> 01:11:55,700
around your claim--

1558
01:11:55,700 --> 01:11:58,180
but do I have a reason to
trust this insurance company

1559
01:11:58,180 --> 01:11:59,972
that, in fact, it's
going to be doing this?

1560
01:11:59,972 --> 01:12:02,680
Then there's some transparency
that this enables--

1561
01:12:02,680 --> 01:12:05,200
not necessarily the transparency
of your credit card number

1562
01:12:05,200 --> 01:12:07,660
or your name or the
flights that you're taking,

1563
01:12:07,660 --> 01:12:10,870
but that there was somebody who
was on this who had a contract,

1564
01:12:10,870 --> 01:12:13,120
and we then made
this kind of payment.

1565
01:12:13,120 --> 01:12:15,100
That would be a potential
value that he has.

1566
01:12:15,100 --> 01:12:17,830
Now, if there's no value
from that, then you're right.

1567
01:12:17,830 --> 01:12:20,110
The overhead of the
blockchain or public

1568
01:12:20,110 --> 01:12:22,240
or private or
whatever is too high,

1569
01:12:22,240 --> 01:12:24,010
and he just could do it inside.

1570
01:12:24,010 --> 01:12:26,320
And for certain
companies, like if you

1571
01:12:26,320 --> 01:12:29,080
went to a credible insurance
company, the credible insurance

1572
01:12:29,080 --> 01:12:31,880
company, you probably trust
that credible insurance company.

1573
01:12:31,880 --> 01:12:35,110
But again, if you're a not
credible insurance company,

1574
01:12:35,110 --> 01:12:39,020
if you're a startup and you
are out there in the market

1575
01:12:39,020 --> 01:12:40,610
and you're saying,
give me your money

1576
01:12:40,610 --> 01:12:42,050
and I'll pay you if
the plane's late,

1577
01:12:42,050 --> 01:12:44,050
people are going to say,
why should I trust you?

1578
01:12:44,050 --> 01:12:45,290
And you're like, well, here.

1579
01:12:45,290 --> 01:12:46,707
You don't really
have to trust me.

1580
01:12:46,707 --> 01:12:48,590
Here's the data.

1581
01:12:48,590 --> 01:12:51,710
Here's the evidence in a more
credible way than if I just

1582
01:12:51,710 --> 01:12:53,900
published a report on
my website that I'm

1583
01:12:53,900 --> 01:12:55,760
doing what I say I'm doing.

1584
01:12:55,760 --> 01:12:56,983
PROFESSOR: Let's try to--

1585
01:12:56,983 --> 01:12:59,793
AUDIENCE: He has
to escrow the $500.

1586
01:12:59,793 --> 01:13:02,210
PROFESSOR: Andy, let's try to
get a couple more questions.

1587
01:13:02,210 --> 01:13:05,570
But my hunch is that it's
probably a permission rather

1588
01:13:05,570 --> 01:13:07,130
than permissionless blockchain.

1589
01:13:07,130 --> 01:13:09,380
And it may be the
insurance company

1590
01:13:09,380 --> 01:13:12,020
wants to ride on the
buzz of blockchain.

1591
01:13:12,020 --> 01:13:15,140
But it's possible,
also, that he thinks

1592
01:13:15,140 --> 01:13:17,930
that there'll be more
trust, that consumers

1593
01:13:17,930 --> 01:13:21,560
will trust that if it's
called a blockchain.

1594
01:13:21,560 --> 01:13:23,150
But you did an interview.

1595
01:13:23,150 --> 01:13:24,580
You can pick who you want to--

1596
01:13:24,580 --> 01:13:25,710
LARRY LESSIG: Right here.

1597
01:13:25,710 --> 01:13:27,127
AUDIENCE: So proof
of performance,

1598
01:13:27,127 --> 01:13:30,420
could this whole smart
contract solve it?

1599
01:13:30,420 --> 01:13:33,350
So for instance, some
countries will receive--

1600
01:13:33,350 --> 01:13:38,960
and I was talking to
Leonardo about it in Cargill,

1601
01:13:38,960 --> 01:13:40,610
in his company.

1602
01:13:40,610 --> 01:13:42,950
They deliver wheat
to some countries.

1603
01:13:42,950 --> 01:13:46,340
And proof of performance
is always delayed.

1604
01:13:46,340 --> 01:13:50,030
And hence, it becomes
a private issue.

1605
01:13:50,030 --> 01:13:53,288
So could smart contracts
solve something like that?

1606
01:13:53,288 --> 01:13:54,080
LARRY LESSIG: Yeah.

1607
01:13:54,080 --> 01:13:58,820
If you had an alternative way
to credibly represent a fact--

1608
01:13:58,820 --> 01:14:02,540
and again, this is kind of
connected to Andy's point.

1609
01:14:02,540 --> 01:14:04,100
Is there an alternative way?

1610
01:14:04,100 --> 01:14:06,910
And is this the alternative way?

1611
01:14:06,910 --> 01:14:10,640
Then you lower the transaction
cost of that kind of contract,

1612
01:14:10,640 --> 01:14:15,560
because I'm going to trust
the contracting process more

1613
01:14:15,560 --> 01:14:20,750
if I'm confident that, in fact,
the fact that I'm not late

1614
01:14:20,750 --> 01:14:22,250
doesn't penalize
me, or you don't

1615
01:14:22,250 --> 01:14:23,750
penalize me claiming I'm late.

1616
01:14:23,750 --> 01:14:26,870
So to the extent you lower
the cost of verifying

1617
01:14:26,870 --> 01:14:29,750
facts in the world, you
increase the opportunity

1618
01:14:29,750 --> 01:14:33,230
for certain contracts to happen
who otherwise wouldn't happen.

1619
01:14:33,230 --> 01:14:35,360
And that's the only
trade-off that I'm

1620
01:14:35,360 --> 01:14:37,550
trying to emphasize by
emphasizing the transaction

1621
01:14:37,550 --> 01:14:39,510
cost part of these contracts.

1622
01:14:39,510 --> 01:14:42,820
You had a question here?

1623
01:14:42,820 --> 01:14:44,368
AUDIENCE: Oh, yeah.

1624
01:14:44,368 --> 01:14:46,660
I was just saying maybe one
of the benefits of applying

1625
01:14:46,660 --> 01:14:51,175
blockchain here would be
not for one-off transactions

1626
01:14:51,175 --> 01:14:52,800
but for the insurance
company, perhaps,

1627
01:14:52,800 --> 01:14:55,537
to have an aggregate
thousands of transactions

1628
01:14:55,537 --> 01:14:56,870
where they've shown performance.

1629
01:14:56,870 --> 01:15:00,070
And they can use this
for commercial reasons

1630
01:15:00,070 --> 01:15:02,330
to show that, yes,
people get paid

1631
01:15:02,330 --> 01:15:05,570
when they purchase our
insurance and get credibility

1632
01:15:05,570 --> 01:15:08,938
for particular products.

1633
01:15:08,938 --> 01:15:09,730
LARRY LESSIG: Yeah.

1634
01:15:09,730 --> 01:15:11,540
You would frame that
as it would have

1635
01:15:11,540 --> 01:15:13,880
to be that that would be
a value it was providing.

1636
01:15:13,880 --> 01:15:15,350
Otherwise, to
Andy's point, there

1637
01:15:15,350 --> 01:15:18,307
would be no reason
to adopt this.

1638
01:15:18,307 --> 01:15:18,890
PROFESSOR: OK.

1639
01:15:18,890 --> 01:15:21,017
Go on the back right.

1640
01:15:21,017 --> 01:15:22,100
I don't remember your name

1641
01:15:22,100 --> 01:15:24,200
AUDIENCE: Kyle.

1642
01:15:24,200 --> 01:15:31,190
So with, I guess, easier access
to these derivative contracts,

1643
01:15:31,190 --> 01:15:33,920
do you foresee a
world where there's

1644
01:15:33,920 --> 01:15:37,992
a lot more leverage
in any type of market

1645
01:15:37,992 --> 01:15:39,200
outside of financial markets?

1646
01:15:39,200 --> 01:15:43,152
And do you foresee this creating
kind of a risk in that way?

1647
01:15:43,152 --> 01:15:46,280
LARRY LESSIG: Well, I mean,
this is your expertise, right?

1648
01:15:46,280 --> 01:15:49,070
Because, I mean, we already
saw in the derivatives market

1649
01:15:49,070 --> 01:15:51,890
the fact that the
choice by policymakers

1650
01:15:51,890 --> 01:15:54,560
was to allow it to be
an invisible market

1651
01:15:54,560 --> 01:15:56,150
and not to regulate that market.

1652
01:15:56,150 --> 01:16:00,487
It created all sorts of risks
that sensible people might

1653
01:16:00,487 --> 01:16:01,820
think shouldn't have been there.

1654
01:16:01,820 --> 01:16:03,195
And I think the
same potential is

1655
01:16:03,195 --> 01:16:05,480
here, which is, again, to
come back to the point,

1656
01:16:05,480 --> 01:16:07,680
don't expect there not
to be government here.

1657
01:16:07,680 --> 01:16:10,910
I mean, to the extent
people realize this risk

1658
01:16:10,910 --> 01:16:13,850
and don't have as much
political corrupting power

1659
01:16:13,850 --> 01:16:15,880
as Wall Street
did in our system,

1660
01:16:15,880 --> 01:16:17,630
we should expect
governments [INAUDIBLE]..

1661
01:16:17,630 --> 01:16:19,640
PROFESSOR: I would say yes.

1662
01:16:19,640 --> 01:16:23,300
Even on bitcoin, you can create
something called discrete log

1663
01:16:23,300 --> 01:16:28,460
contracts to do contracts for
differences or derivatives.

1664
01:16:28,460 --> 01:16:31,280
But with smart contracts,
you can put a lot of leverage

1665
01:16:31,280 --> 01:16:33,880
into the system.

1666
01:16:33,880 --> 01:16:35,770
And transactions
can move quickly.

1667
01:16:35,770 --> 01:16:39,770
So I think the state will
have an interest, particularly

1668
01:16:39,770 --> 01:16:41,120
if it grows.

1669
01:16:41,120 --> 01:16:43,370
We probably only have time
for two more questions.

1670
01:16:43,370 --> 01:16:46,490
And I'm still wondering,
how do courts currently look

1671
01:16:46,490 --> 01:16:50,670
at these as a matter of law?

1672
01:16:50,670 --> 01:16:53,820
In the late 1990s, I
was honored to work

1673
01:16:53,820 --> 01:16:56,190
with Senator John McCain
on something which

1674
01:16:56,190 --> 01:16:58,622
became the e-signature law.

1675
01:16:58,622 --> 01:17:01,080
I was just playing point for
the US Department of Treasury.

1676
01:17:01,080 --> 01:17:05,160
It was really John McCain's
law Congress was doing.

1677
01:17:05,160 --> 01:17:06,900
But I'm wondering
how the courts today

1678
01:17:06,900 --> 01:17:09,720
would see these contracts.

1679
01:17:09,720 --> 01:17:12,840
Do we need something
like an e-signature law

1680
01:17:12,840 --> 01:17:18,642
to have code accepted beyond
what e-signature accepted?

1681
01:17:18,642 --> 01:17:19,350
LARRY LESSIG: No.

1682
01:17:19,350 --> 01:17:22,170
I think the e-signature
infrastructure

1683
01:17:22,170 --> 01:17:24,487
is going to be 99% of
what this system is.

1684
01:17:24,487 --> 01:17:26,820
PROFESSOR: You mean that which
I worked on 20 years ago?

1685
01:17:26,820 --> 01:17:26,912
LARRY LESSIG: Yeah.

1686
01:17:26,912 --> 01:17:28,410
You solved 99% of the problem.

1687
01:17:28,410 --> 01:17:29,618
PROFESSOR: Little did I know.

1688
01:17:29,618 --> 01:17:30,640
LARRY LESSIG: Here, too.

1689
01:17:30,640 --> 01:17:31,140
Yeah.

1690
01:17:31,140 --> 01:17:34,140
I mean, that's why I
wanted to wrap this

1691
01:17:34,140 --> 01:17:36,840
around the idea of a
soda pop machine, right?

1692
01:17:36,840 --> 01:17:39,610
So it's nothing
new, in some sense.

1693
01:17:39,610 --> 01:17:41,610
Contract law has always
dealt with machines that

1694
01:17:41,610 --> 01:17:44,290
are delivering on obligations.

1695
01:17:44,290 --> 01:17:45,960
It's just a more
sophisticated set

1696
01:17:45,960 --> 01:17:48,540
of proofs you're going to
have to make in a court.

1697
01:17:48,540 --> 01:17:52,410
But it's the same
underlying plot.

1698
01:17:52,410 --> 01:17:55,700
PROFESSOR: One more question.

1699
01:17:55,700 --> 01:17:56,280
We're MIT.

1700
01:17:56,280 --> 01:17:57,870
We're supposed to end
five minutes early.

1701
01:17:57,870 --> 01:17:58,260
LARRY LESSIG: Oh, OK.

1702
01:17:58,260 --> 01:17:59,160
PROFESSOR: I don't know
what you do at Harvard.

1703
01:17:59,160 --> 01:18:01,450
LARRY LESSIG: I don't remember.

1704
01:18:01,450 --> 01:18:03,180
PROFESSOR: It's Harvard Law.

1705
01:18:03,180 --> 01:18:05,500
LARRY LESSIG: Right here.

1706
01:18:05,500 --> 01:18:08,460
AUDIENCE: So I have a little
bit of the opposite question

1707
01:18:08,460 --> 01:18:12,660
from Elon, which is a lot of the
contracts you're talking about,

1708
01:18:12,660 --> 01:18:14,220
simple contracts, right?

1709
01:18:14,220 --> 01:18:16,200
You've got individuals
on one side.

1710
01:18:16,200 --> 01:18:18,620
You've got large companies,
like your insurance company,

1711
01:18:18,620 --> 01:18:19,570
on the other.

1712
01:18:19,570 --> 01:18:21,610
The insurance company
can't, in fact,

1713
01:18:21,610 --> 01:18:24,210
go through the
entire decision tree

1714
01:18:24,210 --> 01:18:28,150
and decide what every
term is in their favor.

1715
01:18:28,150 --> 01:18:30,530
And the law deals with
that all the time.

1716
01:18:30,530 --> 01:18:33,230
But here, you have another
problem, which is now,

1717
01:18:33,230 --> 01:18:34,410
it's going to be OK.

1718
01:18:34,410 --> 01:18:36,770
It's going to be written
in computer code.

1719
01:18:36,770 --> 01:18:39,140
How are the judges
going to deal with that?

1720
01:18:39,140 --> 01:18:41,630
Now it's a language
that is going

1721
01:18:41,630 --> 01:18:44,300
to be even tougher to
deal with that problem.

1722
01:18:44,300 --> 01:18:45,930
LARRY LESSIG: Great question.

1723
01:18:45,930 --> 01:18:47,282
And the answer is--

1724
01:18:47,282 --> 01:18:49,490
so this is actually another
version of your question.

1725
01:18:49,490 --> 01:18:51,890
The answer to that is
nobody knows right now.

1726
01:18:51,890 --> 01:18:55,250
And so this issue gets raised
in a lot of parallel contexts.

1727
01:18:55,250 --> 01:18:59,390
For example, there's a Supreme
Court Justice from California

1728
01:18:59,390 --> 01:19:04,880
who's doing a lot of work about
the question of black boxes

1729
01:19:04,880 --> 01:19:09,495
that predict whether you are
likely to commit a crime again.

1730
01:19:09,495 --> 01:19:11,370
So if you're convicted
of crime, the question

1731
01:19:11,370 --> 01:19:12,995
is whether you're
going to have parole.

1732
01:19:12,995 --> 01:19:15,840
The parole process is supposed
to decide your likelihood

1733
01:19:15,840 --> 01:19:17,520
of committing a crime.

1734
01:19:17,520 --> 01:19:20,010
So these companies have
built these black boxes where

1735
01:19:20,010 --> 01:19:21,690
you spit in every bit of data.

1736
01:19:21,690 --> 01:19:24,060
And it comes out and says
yes, you're a criminal,

1737
01:19:24,060 --> 01:19:25,553
or no, you're not.

1738
01:19:25,553 --> 01:19:26,970
And then the lawyers
come forward,

1739
01:19:26,970 --> 01:19:29,640
and they say, well, I want
to interrogate the black box.

1740
01:19:29,640 --> 01:19:31,040
I want to know what--

1741
01:19:31,040 --> 01:19:33,773
And the courts are completely
confused about this.

1742
01:19:33,773 --> 01:19:35,190
Because on the one
hand, they say,

1743
01:19:35,190 --> 01:19:37,725
no, no, this is
their trade secret.

1744
01:19:37,725 --> 01:19:39,850
But on the other hand,
they're like, wait a minute.

1745
01:19:39,850 --> 01:19:41,830
You can't decide whether
I have liberty or not

1746
01:19:41,830 --> 01:19:45,048
based on these black boxes that
just spit out numbers, right?

1747
01:19:45,048 --> 01:19:46,840
So I've got to be able
to interrogate them.

1748
01:19:46,840 --> 01:19:47,710
Right.

1749
01:19:47,710 --> 01:19:51,100
So this is the most
important innovation that's

1750
01:19:51,100 --> 01:19:52,600
got to happen in the law.

1751
01:19:52,600 --> 01:19:55,270
Lawyers have to
become technically

1752
01:19:55,270 --> 01:19:59,140
sophisticated to be
able to read code

1753
01:19:59,140 --> 01:20:01,630
like they read legal contracts.

1754
01:20:01,630 --> 01:20:04,210
They have to have a conceptual
understanding of what

1755
01:20:04,210 --> 01:20:07,990
code can do so they can
think creatively about code

1756
01:20:07,990 --> 01:20:10,960
like they think creatively
about language, right?

1757
01:20:10,960 --> 01:20:13,900
And one of the most
important innovations

1758
01:20:13,900 --> 01:20:17,710
that we, like the Berkman
Center and people like that,

1759
01:20:17,710 --> 01:20:19,870
are trying to push into
the legal environment

1760
01:20:19,870 --> 01:20:23,680
is to insist that just
like lawyers now all study

1761
01:20:23,680 --> 01:20:27,280
economics, because that's a core
way of thinking about the law,

1762
01:20:27,280 --> 01:20:30,480
we also should study
basic code so we

1763
01:20:30,480 --> 01:20:33,910
can think about this
modality of regulation.

1764
01:20:33,910 --> 01:20:36,070
Not expect people are
going to be writing code,

1765
01:20:36,070 --> 01:20:38,230
but at least that they
can think through it.

1766
01:20:38,230 --> 01:20:39,280
So I completely agree.

1767
01:20:39,280 --> 01:20:42,750
That's a fudge space that
I didn't talk about at all.

1768
01:20:42,750 --> 01:20:45,310
But I think it's a really
important problem, thinking

1769
01:20:45,310 --> 01:20:46,540
about whether the
legal system is going

1770
01:20:46,540 --> 01:20:47,665
to be able to deal with it.

1771
01:20:47,665 --> 01:20:51,340
Because if none of the
lawyers understand it, yeah.

1772
01:20:51,340 --> 01:20:53,090
Let me just end with
one story about that.

1773
01:20:53,090 --> 01:20:55,990
So when I first started
doing this work, talking

1774
01:20:55,990 --> 01:20:58,150
about code and--

1775
01:20:58,150 --> 01:21:01,620
I went to Paris 20 years
ago, and I gave a speech

1776
01:21:01,620 --> 01:21:02,650
to some lawyer group.

1777
01:21:02,650 --> 01:21:04,810
And I sat down after the speech.

1778
01:21:04,810 --> 01:21:09,160
And the chairman of this legal
group sat down next to me.

1779
01:21:09,160 --> 01:21:10,210
And I introduced myself.

1780
01:21:10,210 --> 01:21:12,040
And he stood up, and
he grabbed my card.

1781
01:21:12,040 --> 01:21:13,580
And he crushed it.

1782
01:21:13,580 --> 01:21:15,430
And he threw it to the ground.

1783
01:21:15,430 --> 01:21:18,650
And he said, I am
not a technician.

1784
01:21:18,650 --> 01:21:20,750
And his point was
he, as a lawyer,

1785
01:21:20,750 --> 01:21:25,310
could never accept the idea that
what he needed to understand

1786
01:21:25,310 --> 01:21:27,150
was code.

1787
01:21:27,150 --> 01:21:28,680
That was forbidden.

1788
01:21:28,680 --> 01:21:30,530
It was beneath him.

1789
01:21:30,530 --> 01:21:34,950
And I think it's a hugely
important cultural problem

1790
01:21:34,950 --> 01:21:37,160
to integrate this
form of knowledge.

1791
01:21:37,160 --> 01:21:39,650
Because if we don't, then the
problem you're talking about

1792
01:21:39,650 --> 01:21:42,080
is really a huge problem.

1793
01:21:42,080 --> 01:21:44,130
But I think we can, because
we've seen lots of--

1794
01:21:44,130 --> 01:21:46,100
PROFESSOR: So I
want to thank you

1795
01:21:46,100 --> 01:21:50,030
on behalf of the class,
Blockchain and Money.

1796
01:21:50,030 --> 01:21:53,404
[APPLAUSE]

1797
01:21:55,340 --> 01:22:00,230
I also want to say you're much
better than Christopher Lloyd.

1798
01:22:00,230 --> 01:22:02,360
And if anybody wants
to know the joke,

1799
01:22:02,360 --> 01:22:07,640
watch that West Wing episode
that Christopher Lloyd plays

1800
01:22:07,640 --> 01:22:08,960
Professor Lawrence Lessig.

1801
01:22:08,960 --> 01:22:11,000
So thank you.

1802
01:22:11,000 --> 01:22:13,030
See you on Thursday.