1
00:00:01,080 --> 00:00:05,160
Hi everybody and welcome back to
the Elon Musk podcast where we 

2
00:00:05,160 --> 00:00:09,480
discuss all things technology, 
space, entrepreneurship and 

3
00:00:09,480 --> 00:00:11,480
more. 
And today we're talking about 

4
00:00:11,480 --> 00:00:13,560
something that's definitely 
caught the attention of Wall 

5
00:00:13,560 --> 00:00:17,280
Street Tesla shareholders and a 
I enthusiasts alike. 

6
00:00:17,680 --> 00:00:21,920
Tesla stock is climbing higher 
is something to do with the very

7
00:00:21,920 --> 00:00:26,440
own supercomputer at Tesla Dojo.
Recently, Morgan Stanley 

8
00:00:26,440 --> 00:00:30,400
upgraded Tesla stock from equal 
weight to overweight and hiked 

9
00:00:30,400 --> 00:00:34,440
the price target from $200.00 to
$400.00. 

10
00:00:34,600 --> 00:00:38,120
But this optimistic outlook 
largely stems from their view of

11
00:00:38,120 --> 00:00:42,360
Dojo's immense potential, not 
just in advancing Tesla's A I 

12
00:00:42,360 --> 00:00:46,480
capabilities, but also as a 
service that could be offered on

13
00:00:46,480 --> 00:00:48,890
its own. 
Now, Dojo's goal is to 

14
00:00:48,890 --> 00:00:53,290
effectively process millions of 
terabytes of video data captured

15
00:00:53,290 --> 00:00:57,770
from real life driving situation
from Tesla's plus 4 million 

16
00:00:57,770 --> 00:00:59,610
cars. 
Now this goal led to a 

17
00:00:59,610 --> 00:01:02,570
considerably different 
architecture than conventional 

18
00:01:02,570 --> 00:01:07,650
supercomputer designs and going 
live in July 2023, Dojo serves 

19
00:01:07,650 --> 00:01:10,410
as the training ground for 
Tesla's machine learning models 

20
00:01:10,770 --> 00:01:14,210
aimed at bolstering its full 
self driving system or FSD. 

21
00:01:14,600 --> 00:01:17,440
And unlike conventional 
supercomputers, Dojo is 

22
00:01:17,440 --> 00:01:20,920
architected to handle millions 
of terabytes of real life video 

23
00:01:20,920 --> 00:01:25,800
data from over 4 million Tesla 
vehicles on the road right now. 

24
00:01:26,240 --> 00:01:29,520
Now this unique task called for 
an architecture that broke away 

25
00:01:29,520 --> 00:01:33,320
from traditional supercomputer 
designs, focusing on efficiency 

26
00:01:33,320 --> 00:01:36,400
and scalability. 
Now, before Dojo, Tesla was 

27
00:01:36,400 --> 00:01:38,200
already a formidable name in 
computing. 

28
00:01:38,670 --> 00:01:41,910
The company had several parallel
computing clusters, one of which

29
00:01:41,910 --> 00:01:45,350
was, by some metrics, among the 
top supercomputers globally. 

30
00:01:45,670 --> 00:01:48,910
However, there were debates 
about its actual performance due

31
00:01:48,910 --> 00:01:52,910
to unclear metrics. 
These test the GPU clusters were

32
00:01:52,910 --> 00:01:55,830
pivotal in training neural 
networks to better understand 

33
00:01:55,830 --> 00:01:59,880
real world driving conditions, 
and with MOJO, Tesla aims to 

34
00:01:59,880 --> 00:02:02,040
supersede its own lofty 
standards. 

35
00:02:02,240 --> 00:02:07,080
Mentioned first in 2019 during 
Tesla's Autonomy Investors Day, 

36
00:02:07,120 --> 00:02:11,440
MOJO faced some blaze due to 
power and thermal challenges. 

37
00:02:11,800 --> 00:02:15,600
Nevertheless, it officially came
into the limelight in 2021 and 

38
00:02:15,600 --> 00:02:20,640
moved into production use in 
2023, equipped with 10,000 

39
00:02:20,800 --> 00:02:24,240
NVIDIA H100GP U's to enhance his
performance. 

40
00:02:24,750 --> 00:02:27,870
Now Dojo's reception among 
analysts is mixed, yet mostly 

41
00:02:27,870 --> 00:02:29,670
positive. 
And while some say it's 

42
00:02:29,670 --> 00:02:33,070
impressive but won't change the 
supercomputing game, others call

43
00:02:33,070 --> 00:02:36,750
it an inhouse developed game 
changer that can massively speed

44
00:02:36,750 --> 00:02:38,670
up the development of autonomous
vehicles. 

45
00:02:39,070 --> 00:02:41,430
Now the endorsement from Morgan 
Stanley is particularly 

46
00:02:41,430 --> 00:02:45,110
significant, suggesting that 
Dojo could be a catalyst for an 

47
00:02:45,110 --> 00:02:47,790
unprecedented increase in 
Tesla's market value. 

48
00:02:48,230 --> 00:02:52,150
The idea is that Tesla's focus 
on custom designed a I chips for

49
00:02:52,150 --> 00:02:55,660
video training could afford a 
formidable advantage in the 

50
00:02:55,660 --> 00:02:58,100
autonomous vehicle sector and 
beyond. 

51
00:02:59,220 --> 00:03:04,500
At Tesla's A I day in 2021, the 
introduced Dojo, but at the time

52
00:03:04,740 --> 00:03:07,740
they were just getting started. 
They had a first chip and some 

53
00:03:07,740 --> 00:03:11,140
training tiles, but we're still 
in the process of constructing a

54
00:03:11,140 --> 00:03:15,420
full Dojo cabinet in a cluster. 
Fast forward to a I day 22. 

55
00:03:15,980 --> 00:03:19,300
They made considerable headway, 
including developing a full 

56
00:03:19,300 --> 00:03:24,230
system tray, and they're aiming 
to ramp it up to a 100 exaflop 

57
00:03:24,230 --> 00:03:29,110
capability by the end of 2024. 
This is particularly significant

58
00:03:29,150 --> 00:03:32,990
as Tesla's ambitions becoming 
increasingly computer 

59
00:03:32,990 --> 00:03:35,630
constrained, especially in 
selfdriving endeavors. 

60
00:03:35,870 --> 00:03:39,350
Moreover, Tesla's feature 
doesn't just rest on car sales. 

61
00:03:39,670 --> 00:03:42,310
Supercharging network with 
partnerships like the one 

62
00:03:42,510 --> 00:03:45,550
they've received established 
with Hilton Hotels presents 

63
00:03:45,590 --> 00:03:47,270
another potential revenue 
stream. 

64
00:03:47,770 --> 00:03:51,370
Adam Jonas believes that the 
advantages and the advancements 

65
00:03:51,370 --> 00:03:54,850
and chip technology will add to 
this by possibly extending the 

66
00:03:54,850 --> 00:03:58,690
self driving licensing and 
operating system licensing. 

67
00:03:59,050 --> 00:04:01,610
Now let's talk numbers for those
who love them. 

68
00:04:01,890 --> 00:04:05,610
Morgan Stanley's price target 
shift from 2:50 to 400 isn't 

69
00:04:05,610 --> 00:04:07,970
trivial. 
The analyst broke down the value

70
00:04:07,970 --> 00:04:11,490
and factored in Dojo's impact on
the overall valuation. 

71
00:04:11,930 --> 00:04:15,090
This was a pivotal enough shift 
to lift Tesla stock by nearly 

72
00:04:15,090 --> 00:04:19,450
10% following the notes release.
And get this, aside from Tesla, 

73
00:04:19,690 --> 00:04:23,650
Morgan Stanley also sees the 
possibility of non Tesla fleet 

74
00:04:23,650 --> 00:04:26,650
licensing revenue, which means 
the reach of Tesla's 

75
00:04:26,650 --> 00:04:30,890
technological advancements could
go beyond its own lineup of 

76
00:04:30,890 --> 00:04:33,450
electric vehicles. 
And while the core business of 

77
00:04:33,450 --> 00:04:36,650
Tesla has always been tied to 
electric vehicles, this new 

78
00:04:36,650 --> 00:04:40,770
focus on Dojo is a revenue 
generator isn't really a wild 

79
00:04:40,770 --> 00:04:42,570
shift. 
In fact, we've mentioned this 

80
00:04:42,570 --> 00:04:44,690
before. 
The Dojo could potentially be 

81
00:04:44,690 --> 00:04:48,420
offered as a service, 
argumenting its utility and 

82
00:04:48,420 --> 00:04:50,900
contributing to Tesla's overall 
value. 

83
00:04:51,340 --> 00:04:54,180
So the concept isn't new. 
But what is notable is how 

84
00:04:54,180 --> 00:04:58,380
mainstream financial analysts 
are now seeing Dojo not just as 

85
00:04:58,380 --> 00:05:01,900
a little side project for Tesla 
for their own vehicles, but as a

86
00:05:01,900 --> 00:05:05,420
key aspect of Tesla's future 
growth and the value. 

87
00:05:05,740 --> 00:05:07,420
Now. 
One more thing before we go 

88
00:05:07,880 --> 00:05:09,840
There are other market forces at
play. 

89
00:05:09,960 --> 00:05:13,200
Currently, there's a looming 
potential of a massive UAW 

90
00:05:13,200 --> 00:05:16,280
strike that can negatively 
affect Tesla's competitors. 

91
00:05:16,560 --> 00:05:20,680
This labor strike in competing 
companies could inadvertently 

92
00:05:20,680 --> 00:05:25,480
give Tesla a huge edge, further 
fueling optimism around Tesla 

93
00:05:25,480 --> 00:05:29,440
stock and future prospects. 
That wraps up the show today. 

94
00:05:29,640 --> 00:05:32,520
Hope you found it informative. 
If you enjoyed the show, don't 

95
00:05:32,520 --> 00:05:34,680
forget to hit the subscribe or 
the follow button on your 

96
00:05:34,680 --> 00:05:38,300
preferred podcast platform. 
It's absolutely free, takes a 

97
00:05:38,300 --> 00:05:41,060
second, and helps us out 
tremendously, and this way 

98
00:05:41,060 --> 00:05:44,380
you'll never miss out on any 
Elon Musk news. 

99
00:05:44,780 --> 00:05:47,420
O thanks for tuning in and 
please take care of yourselves 

100
00:05:47,420 --> 00:05:49,460
and each other and I'll see you 
in the next one.