WEBVTT

1
00:00:00.080 --> 00:00:03.839
<v Speaker 1>I suspect that the long term bad for JERUMP consumed

2
00:00:03.879 --> 00:00:06.759
<v Speaker 1>right robots is an excess of twenty billion units. So

3
00:00:06.799 --> 00:00:09.919
<v Speaker 1>we're ramping out production in our US factory as well

4
00:00:09.960 --> 00:00:13.480
<v Speaker 1>as building our building megaback factory in China that's such

5
00:00:13.519 --> 00:00:16.960
<v Speaker 1>a roughly double hour output, maybe troubled potentially inclusion.

6
00:00:17.000 --> 00:00:19.399
<v Speaker 2>We're super excited about the progress across the board.

7
00:00:19.519 --> 00:00:21.800
<v Speaker 1>I recommend anyone who doesn't believe that says the will

8
00:00:21.839 --> 00:00:24.839
<v Speaker 1>soult bigle autonomy should not hold test the stop. They

9
00:00:24.839 --> 00:00:28.039
<v Speaker 1>should sell a TESTA stop you should believe will solve autonomy.

10
00:00:28.120 --> 00:00:29.199
<v Speaker 2>You should buy test to stop.

11
00:00:29.359 --> 00:00:32.799
<v Speaker 1>Enabling the fleet to operate like the Giant Autonomous fleet,

12
00:00:33.039 --> 00:00:36.159
<v Speaker 1>and it takes evaluation, I think to some pretty crazy

13
00:00:36.240 --> 00:00:38.600
<v Speaker 1>number of ARC invest thinks on the order of five

14
00:00:38.640 --> 00:00:40.920
<v Speaker 1>trillion dollars. I think they're probably not wrong, and long

15
00:00:41.000 --> 00:00:44.759
<v Speaker 1>term optimists, I think IS achieves evaluation several times that number.

16
00:00:44.880 --> 00:00:47.719
<v Speaker 1>So we'll large adoption acceleration evs and then a bit

17
00:00:47.719 --> 00:00:49.640
<v Speaker 1>of a hangover as other struggle to.

18
00:00:49.640 --> 00:00:50.840
<v Speaker 2>Make compelling evs.

19
00:00:51.039 --> 00:00:53.600
<v Speaker 1>So there's been quite a few competing electric bicles that

20
00:00:53.600 --> 00:00:56.159
<v Speaker 1>have entered the market, and mostly they have not done well,

21
00:00:56.200 --> 00:00:59.039
<v Speaker 1>but they have discounted their quite substantially, which has made

22
00:00:59.079 --> 00:01:01.880
<v Speaker 1>it but the board for Tesla, we don't see this

23
00:01:01.920 --> 00:01:02.799
<v Speaker 1>as a long term issue.

24
00:01:02.840 --> 00:01:04.799
<v Speaker 2>But but really it's a fairly short.

25
00:01:04.680 --> 00:01:07.920
<v Speaker 1>Term and we still obviously firmly believe that evs are

26
00:01:07.959 --> 00:01:11.239
<v Speaker 1>best for customers and that the world is headed for

27
00:01:11.519 --> 00:01:14.560
<v Speaker 1>a fully electrified transport, not just the cars, but also

28
00:01:14.959 --> 00:01:18.120
<v Speaker 1>aircraft and both. Despite many challenges, the Tesla team did

29
00:01:18.120 --> 00:01:19.920
<v Speaker 1>a great job executing.

30
00:01:19.560 --> 00:01:22.280
<v Speaker 2>And we did achieve record quarterly revenues.

31
00:01:22.519 --> 00:01:25.159
<v Speaker 1>Energy storage deployments reached an all time high Q two,

32
00:01:25.319 --> 00:01:28.439
<v Speaker 1>leading to record profits for the energy business, and we're

33
00:01:28.480 --> 00:01:32.200
<v Speaker 1>investing in many future projects, including air training and inference

34
00:01:32.480 --> 00:01:37.120
<v Speaker 1>and great infrastructure to support to future products. We won't

35
00:01:37.159 --> 00:01:39.640
<v Speaker 1>get too much into the product roadmap here because that

36
00:01:39.799 --> 00:01:42.840
<v Speaker 1>is reserved for product announcement events, but we are in

37
00:01:42.879 --> 00:01:45.719
<v Speaker 1>fact to develop to deliver a more affordable model in

38
00:01:45.719 --> 00:01:47.920
<v Speaker 1>the first half of the next year. Really, my father,

39
00:01:47.959 --> 00:01:51.400
<v Speaker 1>the biggest differentiated for Tesla is autonomy. In addition to that,

40
00:01:51.480 --> 00:01:54.400
<v Speaker 1>we have scale economies and we're the most efficient electric

41
00:01:54.439 --> 00:01:58.040
<v Speaker 1>vehicle producer in the world. While while others are pursuing

42
00:01:58.040 --> 00:02:00.680
<v Speaker 1>different parts of the air robotic stac we're presuing all

43
00:02:00.760 --> 00:02:03.079
<v Speaker 1>of them the slows for better cost control, more scale,

44
00:02:03.120 --> 00:02:06.040
<v Speaker 1>cricket time to market, and a superior product applying not

45
00:02:06.120 --> 00:02:09.840
<v Speaker 1>to not just do autonomous to autonomous humanoid robots. Like

46
00:02:09.879 --> 00:02:14.000
<v Speaker 1>optimists regarding full st driving and robot taxi, we've made

47
00:02:14.039 --> 00:02:16.080
<v Speaker 1>a lot of progress with will sub driving in Q two,

48
00:02:16.639 --> 00:02:19.800
<v Speaker 1>and with version twelve point five beginning rollout, we think

49
00:02:19.919 --> 00:02:24.000
<v Speaker 1>customers will experience a staff change improvement in how well

50
00:02:24.199 --> 00:02:27.479
<v Speaker 1>supervised full self driving works. Version twelve point five has

51
00:02:27.639 --> 00:02:30.960
<v Speaker 1>five times the parameters of twelve point four and will

52
00:02:31.039 --> 00:02:34.280
<v Speaker 1>finally merge the highway and city stacks, so the highway

53
00:02:34.280 --> 00:02:34.680
<v Speaker 1>side is.

54
00:02:34.599 --> 00:02:36.039
<v Speaker 2>Soil at this point's pretty old.

55
00:02:36.120 --> 00:02:38.719
<v Speaker 1>So often the issues people like can are on the highway,

56
00:02:38.960 --> 00:02:41.560
<v Speaker 1>but with twelve point five we're finally merged the two stacks.

57
00:02:41.639 --> 00:02:43.960
<v Speaker 1>I still find out that most people actually don't know

58
00:02:43.960 --> 00:02:46.639
<v Speaker 1>how good the system is, and I would encourage anyone

59
00:02:46.800 --> 00:02:49.479
<v Speaker 1>to understand system better to simply try it out and

60
00:02:49.800 --> 00:02:51.439
<v Speaker 1>let the car drive you around. One of the things

61
00:02:51.439 --> 00:02:53.520
<v Speaker 1>we're going to be doing just to make sure people

62
00:02:53.560 --> 00:02:57.240
<v Speaker 1>actually understand the capabilities of the car is when delivering

63
00:02:57.319 --> 00:02:59.599
<v Speaker 1>a new car and when picking up a car for

64
00:02:59.800 --> 00:03:01.960
<v Speaker 1>so is to just show people how to use it

65
00:03:02.080 --> 00:03:04.719
<v Speaker 1>and just drapping around the block. Once people use it

66
00:03:04.759 --> 00:03:07.759
<v Speaker 1>at all, that they tend to continue using it. So

67
00:03:07.800 --> 00:03:10.919
<v Speaker 1>it's it's very compelling, and this I think will be

68
00:03:10.960 --> 00:03:14.360
<v Speaker 1>a massive demand driver. Even unsupervisedable self driving will be

69
00:03:14.560 --> 00:03:17.919
<v Speaker 1>a massive demand driver. And as we increase the miles

70
00:03:17.919 --> 00:03:21.879
<v Speaker 1>between intervention, it will transition from supervisable self driving to

71
00:03:22.520 --> 00:03:26.599
<v Speaker 1>unsupervisable self driving and we can unlock massive potential in

72
00:03:26.719 --> 00:03:30.400
<v Speaker 1>Deeplee were postponed the sort of Robotaxia with the product

73
00:03:30.560 --> 00:03:34.680
<v Speaker 1>unveiled by a couple months, where it's shifted to ten

74
00:03:34.759 --> 00:03:37.719
<v Speaker 1>ten to October, and this is because I wanted to

75
00:03:37.759 --> 00:03:41.240
<v Speaker 1>take some important changes that I think would improve the vehicle,

76
00:03:41.360 --> 00:03:43.479
<v Speaker 1>the sort of the Robotaxi. That think that I think

77
00:03:43.520 --> 00:03:44.879
<v Speaker 1>that we're the main thing that we're going to show

78
00:03:44.960 --> 00:03:46.840
<v Speaker 1>and we're also going to show up a couple of

79
00:03:46.879 --> 00:03:49.199
<v Speaker 1>other things. So moving back, moving it back a few

80
00:03:49.199 --> 00:03:53.400
<v Speaker 1>months allowed us to improve the Robotaxi as well as adding.

81
00:03:53.199 --> 00:03:55.120
<v Speaker 2>A couple of other things the product unveil.

82
00:03:55.159 --> 00:03:58.120
<v Speaker 1>We're also nearing completion of the self expension of Being Texas,

83
00:03:58.159 --> 00:04:01.319
<v Speaker 1>which will has the largest training cluster to day. So

84
00:04:01.439 --> 00:04:05.599
<v Speaker 1>the incremental fifty eight one hundreds plus twenty thousand of

85
00:04:05.599 --> 00:04:09.759
<v Speaker 1>our hardware for or AI cive a Tesla AI computer

86
00:04:09.919 --> 00:04:14.000
<v Speaker 1>with Optimists. Optimis is already forming tasks in our factory

87
00:04:14.159 --> 00:04:18.720
<v Speaker 1>and we expect to have Optimist production version one in

88
00:04:18.759 --> 00:04:21.639
<v Speaker 1>limited production starting early next year. This will be for

89
00:04:22.120 --> 00:04:24.600
<v Speaker 1>Tesla consumption. It's just better for us to iron out

90
00:04:24.600 --> 00:04:28.680
<v Speaker 1>the issues ourselves. But we expect to have several thousand

91
00:04:28.800 --> 00:04:31.759
<v Speaker 1>Optimus robots produced and doing useful things by the end

92
00:04:31.800 --> 00:04:33.879
<v Speaker 1>of next year in the Tesli factories, and then in

93
00:04:33.959 --> 00:04:37.560
<v Speaker 1>twenty twenty six wramping up production quite a bit, and

94
00:04:37.800 --> 00:04:41.680
<v Speaker 1>at that point we'll be providing Optimist robots to outside customers.

95
00:04:41.839 --> 00:04:45.079
<v Speaker 1>WELV production version two of Optimists. For the energy business,

96
00:04:45.240 --> 00:04:48.399
<v Speaker 1>this is growing fast than anything else. This is we

97
00:04:48.480 --> 00:04:51.519
<v Speaker 1>are really demand constrained rather than production constrained. So we're

98
00:04:51.519 --> 00:04:54.680
<v Speaker 1>wramping up production in our US factory as well as

99
00:04:54.879 --> 00:04:58.120
<v Speaker 1>building our building megaback factory in China that's such a

100
00:04:58.319 --> 00:05:01.879
<v Speaker 1>roughly double our output, more than network maybe troubled potentially.

101
00:05:02.000 --> 00:05:04.959
<v Speaker 1>In conclusion, we're super excited about the progress across the board.

102
00:05:05.040 --> 00:05:07.839
<v Speaker 1>We're changing the energy system, how people move around, happy

103
00:05:07.839 --> 00:05:10.240
<v Speaker 1>wople approach the economy, the art taking is massive, but

104
00:05:10.279 --> 00:05:12.480
<v Speaker 1>I think the future is incredibly bright. It really just

105
00:05:12.839 --> 00:05:16.839
<v Speaker 1>can't emphasize just importance of autonomy for the vesical side

106
00:05:16.879 --> 00:05:20.480
<v Speaker 1>and for optimists. Although the numbers sound crazy, I think

107
00:05:20.720 --> 00:05:24.959
<v Speaker 1>Tesla producing app volume with unsupervised MSD essentially the fleet

108
00:05:25.079 --> 00:05:29.439
<v Speaker 1>to operate like a giant autonomous fleet, and it takes evaluation.

109
00:05:29.600 --> 00:05:32.399
<v Speaker 1>I think to some pretty crazy number are invest thinks

110
00:05:32.439 --> 00:05:34.560
<v Speaker 1>on the order of five trillion dollars, I think they're

111
00:05:34.600 --> 00:05:35.279
<v Speaker 1>probably not wrong.

112
00:05:35.399 --> 00:05:36.600
<v Speaker 2>And long term optimists.

113
00:05:36.680 --> 00:05:40.120
<v Speaker 1>Enabling the fleet to operate like the giant autonomous fleet

114
00:05:40.360 --> 00:05:43.480
<v Speaker 1>and it takes evaluation, I think to some pretty crazy

115
00:05:43.560 --> 00:05:45.920
<v Speaker 1>number of ARC invest thinks on the order of five

116
00:05:45.959 --> 00:05:48.240
<v Speaker 1>trillion dollars, I think they're probably not wrong. And long

117
00:05:48.319 --> 00:05:52.120
<v Speaker 1>term optimists, I think is achieves evaluation several times that number.

118
00:05:52.319 --> 00:05:54.480
<v Speaker 1>When can we do on supervidens to full self driving,

119
00:05:54.560 --> 00:05:57.639
<v Speaker 1>It's difficult. My predictions on this have been overly optimists

120
00:05:57.639 --> 00:06:00.480
<v Speaker 1>for the past our trend it seems as though we

121
00:06:00.480 --> 00:06:04.000
<v Speaker 1>should get miles between interventions to be high enough. That

122
00:06:04.480 --> 00:06:06.959
<v Speaker 1>too far enough the access to humans that you could

123
00:06:06.959 --> 00:06:09.600
<v Speaker 1>do untivized, possibly by the end of this year.

124
00:06:09.839 --> 00:06:11.800
<v Speaker 2>I would be shocked if we cannot do it next year.

125
00:06:11.920 --> 00:06:15.000
<v Speaker 1>So nexture seems highly probable to me based on when

126
00:06:15.040 --> 00:06:18.399
<v Speaker 1>simply plus the points in the curve of miles between intervention,

127
00:06:18.680 --> 00:06:22.600
<v Speaker 1>that trend exceeds human pushure next year. I'm quite concerned

128
00:06:22.639 --> 00:06:25.720
<v Speaker 1>about actually being able to get steady art NA gvus

129
00:06:26.040 --> 00:06:29.160
<v Speaker 1>when we want them, and I think this therefore requires

130
00:06:29.199 --> 00:06:31.720
<v Speaker 1>that we put a lot more effort on dojo in

131
00:06:31.800 --> 00:06:34.000
<v Speaker 1>order to have in order to ensure that we've got

132
00:06:34.000 --> 00:06:35.920
<v Speaker 1>the training capability that we need. We are going to

133
00:06:35.920 --> 00:06:38.800
<v Speaker 1>double down on dojo, and we do see a path

134
00:06:38.839 --> 00:06:41.680
<v Speaker 1>to being competitive with the video with dojo, and I

135
00:06:41.720 --> 00:06:44.560
<v Speaker 1>think we kind of have no choice because the demand

136
00:06:44.600 --> 00:06:48.079
<v Speaker 1>for video is so high and that it's obviously their

137
00:06:48.120 --> 00:06:51.639
<v Speaker 1>obligation essentially to raise the price of GPUs to whatever

138
00:06:51.680 --> 00:06:54.040
<v Speaker 1>the market will bear, which is very high.

139
00:06:54.160 --> 00:06:56.399
<v Speaker 2>I think we've really got to make dojo work, and

140
00:06:56.480 --> 00:06:56.800
<v Speaker 2>we will.

141
00:06:57.079 --> 00:07:00.120
<v Speaker 1>Oustmus is intended to be a journalized humanoid robot a

142
00:07:00.160 --> 00:07:02.160
<v Speaker 1>lot of intelligence, So it's like saying, what kind of

143
00:07:02.160 --> 00:07:05.639
<v Speaker 1>accessories where we offered really human It's really intended to

144
00:07:05.639 --> 00:07:09.199
<v Speaker 1>be able to be backwards compatible with human tasks, so

145
00:07:09.360 --> 00:07:11.519
<v Speaker 1>it would be used any accessories that of human reduced.

146
00:07:11.600 --> 00:07:15.040
<v Speaker 1>Trump has said that heavy towers on Bigle's produced in Mexico,

147
00:07:15.160 --> 00:07:17.639
<v Speaker 1>So it doesn't make sense to invest a lot in

148
00:07:18.040 --> 00:07:20.079
<v Speaker 1>Mexico if that is going to be the case. So

149
00:07:20.199 --> 00:07:22.360
<v Speaker 1>we can see where things to play out politically. But

150
00:07:22.439 --> 00:07:26.279
<v Speaker 1>we are increasing in capacity at our existing factories quite significant.

151
00:07:26.319 --> 00:07:29.600
<v Speaker 1>I should say that the cyber taxi or robotaxi will

152
00:07:29.639 --> 00:07:32.839
<v Speaker 1>get produced here at our headquarters that get to Texas, right,

153
00:07:32.920 --> 00:07:36.319
<v Speaker 1>thank you as well, and as well Optimists towards the

154
00:07:36.360 --> 00:07:39.199
<v Speaker 1>end next year for Optimist production version to the high

155
00:07:39.240 --> 00:07:41.600
<v Speaker 1>volume version of Optimists, we're also be produced here in Texas.

156
00:07:41.720 --> 00:07:45.360
<v Speaker 1>I think that there are opportunities to integrate growth into software.

157
00:07:45.839 --> 00:07:48.720
<v Speaker 1>So it was just there was this was this big

158
00:07:48.759 --> 00:07:52.519
<v Speaker 1>accent of a Tesla. There was no sales, the data

159
00:07:52.519 --> 00:07:55.160
<v Speaker 1>centers were full, there was no place to actually put

160
00:07:55.160 --> 00:07:59.120
<v Speaker 1>them be we were. We've been twenty four to seven

161
00:07:59.480 --> 00:08:02.560
<v Speaker 1>to completely the self extension on the Tesla Big factory

162
00:08:02.600 --> 00:08:03.160
<v Speaker 1>here in Texas.

163
00:08:03.319 --> 00:08:06.040
<v Speaker 2>That's self extension is what will house.

164
00:08:05.800 --> 00:08:09.519
<v Speaker 1>The fifty four hundreds and we're beginning to move the

165
00:08:09.759 --> 00:08:12.480
<v Speaker 1>h one hundred server acts each place there. But we

166
00:08:12.480 --> 00:08:15.319
<v Speaker 1>were really needed we needed that to be complete. Physically,

167
00:08:15.600 --> 00:08:18.600
<v Speaker 1>if you can't order computer ordered GPUs and turn them on,

168
00:08:18.800 --> 00:08:20.800
<v Speaker 1>but no, you need a data. Is that it's possible,

169
00:08:20.879 --> 00:08:23.040
<v Speaker 1>don't want to be clear. That was in Tesla's interests,

170
00:08:23.079 --> 00:08:26.040
<v Speaker 1>not country Tesla centries. So the Tesla a no good

171
00:08:26.040 --> 00:08:28.959
<v Speaker 1>to have GPUs that catch enough this, but second that

172
00:08:29.160 --> 00:08:32.200
<v Speaker 1>self extension is able to take the GPUs, which is

173
00:08:32.399 --> 00:08:35.039
<v Speaker 1>really just this week we're moving GPU's in there and

174
00:08:35.080 --> 00:08:35.960
<v Speaker 1>we'll bring them online.

175
00:08:36.039 --> 00:08:40.159
<v Speaker 2>It was about to Xai, the only want.

176
00:08:40.039 --> 00:08:42.639
<v Speaker 1>To work on a GI So what I was finding

177
00:08:43.000 --> 00:08:45.600
<v Speaker 1>was that we're trying to compuble to to Tesla. They

178
00:08:45.639 --> 00:08:49.120
<v Speaker 1>were only interested in working on APM, not testing problems.

179
00:08:49.240 --> 00:08:51.759
<v Speaker 2>And they want to start they do a startup.

180
00:08:51.799 --> 00:08:53.919
<v Speaker 1>So it was case of either they go to a

181
00:08:53.960 --> 00:08:56.639
<v Speaker 1>startup or and I'm involved, or do a startup and.

182
00:08:56.639 --> 00:08:57.320
<v Speaker 2>I'm not involved.

183
00:08:57.399 --> 00:08:59.720
<v Speaker 1>Those are the two choices. This wasn't they were They

184
00:08:59.720 --> 00:09:01.519
<v Speaker 1>were to Tesla. They were not going to come to

185
00:09:01.519 --> 00:09:04.320
<v Speaker 1>Tesla or any circumstances. Yeah, billion people on Earth, so

186
00:09:04.320 --> 00:09:06.799
<v Speaker 1>it's a billion right there. Then you've got all of

187
00:09:06.840 --> 00:09:09.480
<v Speaker 1>the industrial uses, which is probably at least as much.

188
00:09:09.679 --> 00:09:11.000
<v Speaker 2>I's not way more so.

189
00:09:11.039 --> 00:09:14.600
<v Speaker 1>I suspect that the long term evand for general purpose

190
00:09:14.679 --> 00:09:17.840
<v Speaker 1>humanide robots is an excess of twenty billion unit and

191
00:09:18.039 --> 00:09:20.679
<v Speaker 1>Tesla is that has the most advanced humorid robot in

192
00:09:20.720 --> 00:09:23.120
<v Speaker 1>the world, and it is also very good at manufacturing,

193
00:09:23.120 --> 00:09:24.320
<v Speaker 1>which each other companies are not.

194
00:09:24.480 --> 00:09:26.600
<v Speaker 2>And we've got a lot of experience, with.

195
00:09:26.399 --> 00:09:28.679
<v Speaker 1>The most experience where the world leaders in real world

196
00:09:28.720 --> 00:09:32.200
<v Speaker 1>am so we have all of the ingredients. I think

197
00:09:32.240 --> 00:09:36.000
<v Speaker 1>we're unique in having all of the ingredients necessary large scale,

198
00:09:36.480 --> 00:09:40.279
<v Speaker 1>high utility generalized humanoid robots. But that's why my Brough

199
00:09:40.440 --> 00:09:43.840
<v Speaker 1>estimate long term is in accordance with the oar invest

200
00:09:43.919 --> 00:09:46.559
<v Speaker 1>analysis of a mock pap of on the order of

201
00:09:46.559 --> 00:09:50.279
<v Speaker 1>five trillion for maybe more for Commoss transport, and it's

202
00:09:50.320 --> 00:09:53.600
<v Speaker 1>several times that number for general purpose humanide robots.

203
00:09:53.679 --> 00:09:55.200
<v Speaker 2>I mean, at that point, I'm not sure what money

204
00:09:55.200 --> 00:09:56.759
<v Speaker 2>even means. But in the end the.

205
00:09:56.720 --> 00:09:59.559
<v Speaker 1>Benign Ai scenario, we are headed for an age of

206
00:09:59.559 --> 00:10:02.840
<v Speaker 1>abundance where there is no shortage of business services.

207
00:10:03.240 --> 00:10:05.679
<v Speaker 2>Anyone can have pretty much anything. It's a while verty

208
00:10:05.679 --> 00:10:08.960
<v Speaker 2>while future we're edible. The distributed compute, Yeah, just really compute.

209
00:10:09.000 --> 00:10:10.639
<v Speaker 2>That seems like a pretty obvious thing to do.

210
00:10:11.039 --> 00:10:14.639
<v Speaker 1>I think the where the distributed compute becomes in staying

211
00:10:14.799 --> 00:10:17.919
<v Speaker 1>is with the next generation Tesla AI chook, which is

212
00:10:18.279 --> 00:10:21.200
<v Speaker 1>hard of reviable work, which is from the standpoint of

213
00:10:21.240 --> 00:10:25.320
<v Speaker 1>an infurnace capability comparable to B two hundred two hundred

214
00:10:25.360 --> 00:10:27.320
<v Speaker 1>and we're able to have in production at the end

215
00:10:27.320 --> 00:10:29.759
<v Speaker 1>of next year and scale production in twenty six.

216
00:10:30.039 --> 00:10:31.480
<v Speaker 2>So it just.

217
00:10:31.200 --> 00:10:34.120
<v Speaker 1>Seems if you've got even if you've got autonomous vehicles

218
00:10:34.120 --> 00:10:36.360
<v Speaker 1>that are operating four fifty or sixty hours a week,

219
00:10:36.399 --> 00:10:38.519
<v Speaker 1>there's one one hundred and sixty eight hours in a week.

220
00:10:38.600 --> 00:10:42.039
<v Speaker 1>So we have somewhere above I think one hundred neural

221
00:10:42.120 --> 00:10:44.879
<v Speaker 1>nen computing. I think we need a better word, the GPU,

222
00:10:44.879 --> 00:10:48.320
<v Speaker 1>because GPU means craps plus and so there's a higherd

223
00:10:48.320 --> 00:10:51.159
<v Speaker 1>hours plus per week of AI compute AI in burns

224
00:10:51.200 --> 00:10:54.480
<v Speaker 1>compute from the fleet and vehicles and probably some percentage

225
00:10:54.480 --> 00:10:57.120
<v Speaker 1>from human right robots that it would make sense to

226
00:10:57.159 --> 00:10:59.720
<v Speaker 1>do distributed in burns. And if you go there's a

227
00:10:59.720 --> 00:11:02.919
<v Speaker 1>fleet at some point one hundred million vehicles with AI

228
00:11:02.919 --> 00:11:05.840
<v Speaker 1>five and beyond AI six and seven whatnot, and they

229
00:11:05.879 --> 00:11:08.440
<v Speaker 1>may be billions of human or robots.

230
00:11:08.639 --> 00:11:10.759
<v Speaker 2>That is just a staggering.

231
00:11:10.279 --> 00:11:13.840
<v Speaker 1>Amount of inforenced computer or that could be used for

232
00:11:13.919 --> 00:11:15.240
<v Speaker 1>general purpose computing.

233
00:11:15.240 --> 00:11:17.200
<v Speaker 2>It doesn't have to be used for the human drobot

234
00:11:17.320 --> 00:11:17.960
<v Speaker 2>or for the car.

235
00:11:18.200 --> 00:11:21.559
<v Speaker 1>So I think you know that's that's a pretty obvious

236
00:11:21.559 --> 00:11:23.879
<v Speaker 1>thing to say. It's more useful than having to do

237
00:11:23.960 --> 00:11:27.120
<v Speaker 1>nothing with that unique gateway computer too, So it's really

238
00:11:27.120 --> 00:11:29.919
<v Speaker 1>a gateway computer with the cellular and Wi Fi, the

239
00:11:30.279 --> 00:11:33.360
<v Speaker 1>tails like computer and seven cameras report nunt cameras that

240
00:11:33.440 --> 00:11:36.399
<v Speaker 1>get efferency sage review. But this will give it give

241
00:11:36.399 --> 00:11:38.759
<v Speaker 1>them the speed at which the border industry moves. It

242
00:11:38.759 --> 00:11:42.720
<v Speaker 1>would be several years before you receive solution is a

243
00:11:42.759 --> 00:11:45.519
<v Speaker 1>generalized solution. Whatever else has if you see a like

244
00:11:45.679 --> 00:11:47.960
<v Speaker 1>way Moo and what not, that very localized solutions that

245
00:11:48.000 --> 00:11:51.679
<v Speaker 1>requires high density mapping. That's not it's quite fragile. Interability

246
00:11:51.720 --> 00:11:54.480
<v Speaker 1>to expand out, believes is limited. Solution is a general

247
00:11:54.519 --> 00:11:55.639
<v Speaker 1>solution that works anywhere.

248
00:11:55.639 --> 00:11:57.600
<v Speaker 2>It would even work on a different Earth. So if

249
00:11:57.600 --> 00:11:59.679
<v Speaker 2>you're rendered a new Earth, it would work on a

250
00:11:59.720 --> 00:12:00.000
<v Speaker 2>new Earth.

251
00:12:00.320 --> 00:12:05.000
<v Speaker 1>So it's this capability, I think in our experience, once

252
00:12:05.080 --> 00:12:09.000
<v Speaker 1>we demonstrate that something is safe enough or significantly safer

253
00:12:09.039 --> 00:12:12.879
<v Speaker 1>than human, we find that regulators are supportive of deployment.

254
00:12:12.919 --> 00:12:14.240
<v Speaker 2>Deployment of that capability.

255
00:12:14.279 --> 00:12:16.840
<v Speaker 1>It's difficult to argue with if you've got a large

256
00:12:17.039 --> 00:12:19.360
<v Speaker 1>number of If you've got billions of miles that show

257
00:12:19.399 --> 00:12:23.360
<v Speaker 1>that in future unsupervised FSD is safer than human what

258
00:12:23.480 --> 00:12:24.440
<v Speaker 1>regulator could.

259
00:12:24.320 --> 00:12:25.639
<v Speaker 2>Really stand in the way of that.

260
00:12:25.720 --> 00:12:29.919
<v Speaker 1>They're marly obligated to approve, So I don't think regulatory

261
00:12:29.919 --> 00:12:32.440
<v Speaker 1>approval will be a limiting factor. I should aposed to

262
00:12:32.480 --> 00:12:36.000
<v Speaker 1>say that the self driving capabilities of that are deployed

263
00:12:36.000 --> 00:12:39.799
<v Speaker 1>outside of North America are far behind that in North America,

264
00:12:39.960 --> 00:12:44.720
<v Speaker 1>so with the twelve point five and maybe twelve point six,

265
00:12:44.759 --> 00:12:48.440
<v Speaker 1>but pretty soon we will ask for regular regulatory approval

266
00:12:48.559 --> 00:12:54.480
<v Speaker 1>of the Tesla supervised FSD in Europe, China and other countries,

267
00:12:54.639 --> 00:12:56.960
<v Speaker 1>and I think we're likely to receive that before the

268
00:12:57.039 --> 00:12:58.919
<v Speaker 1>end of the year, thank you, which will be a

269
00:12:58.919 --> 00:13:02.840
<v Speaker 1>helpful demand driver in those region. Tops O JAM cancel

270
00:13:02.879 --> 00:13:04.919
<v Speaker 1>it is because JAM can't make it work, not because

271
00:13:04.960 --> 00:13:07.720
<v Speaker 1>of regulators. They're blaming regulators. That's most leading up then

272
00:13:07.879 --> 00:13:10.360
<v Speaker 1>to do because a Waimo is doing just buying in

273
00:13:10.399 --> 00:13:14.000
<v Speaker 1>those buckets. So it's just that their technology is not far.

274
00:13:14.159 --> 00:13:15.879
<v Speaker 1>But there's the sh A number of times I've cooked

275
00:13:15.879 --> 00:13:17.840
<v Speaker 1>with smart people who live in New York or in

276
00:13:18.120 --> 00:13:21.480
<v Speaker 1>downtown Boston and don't ever drive and then are asking

277
00:13:21.519 --> 00:13:23.759
<v Speaker 1>me about FST. I'm like, you can just get a

278
00:13:23.799 --> 00:13:26.600
<v Speaker 1>car and try it, and if you're not doing that.

279
00:13:26.519 --> 00:13:28.559
<v Speaker 2>You have no idea what's going on. This would just

280
00:13:28.600 --> 00:13:29.480
<v Speaker 2>be the Tesla network.

281
00:13:29.480 --> 00:13:31.799
<v Speaker 1>You just literally open the Tesla app and sum in

282
00:13:31.840 --> 00:13:33.639
<v Speaker 1>a car and re sent a car to pick you

283
00:13:33.720 --> 00:13:36.919
<v Speaker 1>up and take you somewhere, and you can. We will

284
00:13:36.919 --> 00:13:39.720
<v Speaker 1>have a fleet that's on the order of seven million,

285
00:13:39.840 --> 00:13:42.000
<v Speaker 1>that a cateful of autonomy. Soon in the US to

286
00:13:42.039 --> 00:13:45.080
<v Speaker 1>come it will be over ten million, then over twenty million.

287
00:13:45.240 --> 00:13:48.320
<v Speaker 1>This is an immense scale and the car is able

288
00:13:48.360 --> 00:13:51.679
<v Speaker 1>to operate twenty four to seven unlike the human driver

289
00:13:51.879 --> 00:13:55.240
<v Speaker 1>the capability to if there's this basically instant scale with

290
00:13:55.360 --> 00:13:59.080
<v Speaker 1>a software. And now this is for a customer and fleet,

291
00:13:59.360 --> 00:14:01.879
<v Speaker 1>so you think of that as being a bit like Airbnb.

292
00:14:02.000 --> 00:14:04.000
<v Speaker 1>You can choose to allow your car to be used

293
00:14:04.000 --> 00:14:06.399
<v Speaker 1>by the fleet or cancel it and bring it back.

294
00:14:06.480 --> 00:14:08.240
<v Speaker 1>You'll be used by the fleet all the time, can

295
00:14:08.279 --> 00:14:09.679
<v Speaker 1>be used by the fleet and some of the time,

296
00:14:09.799 --> 00:14:11.960
<v Speaker 1>and then Tesla would take would share on the revenue

297
00:14:12.039 --> 00:14:13.720
<v Speaker 1>with the customer. But you can think of the giant

298
00:14:13.720 --> 00:14:15.799
<v Speaker 1>fleet of Tesla vehicles as like the giant sort of

299
00:14:15.879 --> 00:14:20.120
<v Speaker 1>Airbnb equivalent fleet, Airbnb on wheels. And in addition, we

300
00:14:20.159 --> 00:14:22.720
<v Speaker 1>would make some number of cars for Tesla that would

301
00:14:22.720 --> 00:14:25.159
<v Speaker 1>just be owned by Tesla and beat that to the fleet.

302
00:14:25.279 --> 00:14:26.679
<v Speaker 2>I guess that would be a bit more like Uber.

303
00:14:26.879 --> 00:14:28.320
<v Speaker 2>But this will be a test a network.

304
00:14:28.399 --> 00:14:31.879
<v Speaker 1>And there's an important clause we've put in every Tesla purchase,

305
00:14:31.919 --> 00:14:34.559
<v Speaker 1>which is that the Tesla vehicles can only be used

306
00:14:34.639 --> 00:14:36.879
<v Speaker 1>in the Tesla fleet. They cannot be used by third

307
00:14:36.879 --> 00:14:39.919
<v Speaker 1>party for or ting. The entire testl of fleet basically

308
00:14:40.159 --> 00:14:42.960
<v Speaker 1>becomes active. And this is obviously maybe the number of

309
00:14:43.000 --> 00:14:45.240
<v Speaker 1>people who don't want their car to own money, but

310
00:14:45.320 --> 00:14:47.639
<v Speaker 1>I think most people will. Yeah, we only find customers

311
00:14:47.639 --> 00:14:50.320
<v Speaker 1>that they can put together a hodge Pludge solution, and

312
00:14:50.360 --> 00:14:53.320
<v Speaker 1>then sometimes they'll pick that solutions and then that doesn't work.

313
00:14:53.559 --> 00:14:55.879
<v Speaker 2>Yeah. I think people don't understand just how much demand

314
00:14:55.919 --> 00:14:57.519
<v Speaker 2>it will be for frig storage.

315
00:14:57.600 --> 00:15:00.679
<v Speaker 1>They really just like the PUP are understating this demand

316
00:15:00.720 --> 00:15:04.519
<v Speaker 1>by product orders magnitude, so that the actual energy total

317
00:15:04.600 --> 00:15:08.120
<v Speaker 1>energy output of state the US grid is if the

318
00:15:08.200 --> 00:15:11.720
<v Speaker 1>power plants can operate at steady state, is at least

319
00:15:11.720 --> 00:15:13.879
<v Speaker 1>to three times the amount of energy it can't be

320
00:15:13.919 --> 00:15:16.360
<v Speaker 1>Producers because there are a huge gap. There's a huge

321
00:15:16.360 --> 00:15:19.919
<v Speaker 1>difference in the from peak to trough in terms of

322
00:15:20.039 --> 00:15:23.960
<v Speaker 1>energy of power generation. In order for a grid to

323
00:15:23.960 --> 00:15:27.080
<v Speaker 1>not have blackouts, it must be able to support the

324
00:15:27.159 --> 00:15:28.919
<v Speaker 1>load at the worst minute of the worst day of

325
00:15:28.960 --> 00:15:31.799
<v Speaker 1>the year, the coldest or hottest stay, which means that

326
00:15:32.000 --> 00:15:33.320
<v Speaker 1>for the rest of the time, for the rest of

327
00:15:33.360 --> 00:15:36.480
<v Speaker 1>the year, it's got massive excess power generation capability, but

328
00:15:36.519 --> 00:15:39.360
<v Speaker 1>it has no way to store that energy. Once you

329
00:15:39.440 --> 00:15:42.679
<v Speaker 1>add battery packs, you can now run the power plants

330
00:15:42.879 --> 00:15:46.200
<v Speaker 1>at steady state. Steady state means that basically any given

331
00:15:46.240 --> 00:15:48.639
<v Speaker 1>grid anywhere in the world can produce in terms of

332
00:15:48.679 --> 00:15:51.000
<v Speaker 1>cumulative energy in the course of the year, at least

333
00:15:51.080 --> 00:15:52.639
<v Speaker 1>twice what it is currently producing.

334
00:15:52.720 --> 00:15:55.440
<v Speaker 2>That's okay, just maybe, I thank you. It's a very

335
00:15:55.440 --> 00:15:56.159
<v Speaker 2>profound thing.

336
00:15:56.519 --> 00:15:58.759
<v Speaker 1>I guess that there would be like some impact, but

337
00:15:58.879 --> 00:16:01.639
<v Speaker 1>I think it would be devastating our competitors and it

338
00:16:01.679 --> 00:16:06.000
<v Speaker 1>would hurt TEXT slightly, but long term probably actually helps Tesla.

339
00:16:05.679 --> 00:16:08.440
<v Speaker 2>It would be my guess. Yeah, but yeah, I said

340
00:16:08.480 --> 00:16:09.440
<v Speaker 2>this before. It's cooled.

341
00:16:10.879 --> 00:16:14.080
<v Speaker 1>If value of Tesla overwhelmingly this autonomy, these other things

342
00:16:14.159 --> 00:16:17.519
<v Speaker 1>are in the noise relative to autonomy, so I recommend

343
00:16:17.559 --> 00:16:20.080
<v Speaker 1>anyone who doesn't believe that says the will soul bigle

344
00:16:20.120 --> 00:16:22.200
<v Speaker 1>autonomy should not hold testa Stop.

345
00:16:22.240 --> 00:16:23.440
<v Speaker 2>They should sell the testa stop.

346
00:16:23.480 --> 00:16:26.080
<v Speaker 1>You should believe Tesla will solve autonomy, you should buy

347
00:16:26.120 --> 00:16:26.679
<v Speaker 1>testa Stop.

348
00:16:26.759 --> 00:16:28.960
<v Speaker 2>And all these other questions are in the noise.