1
00:00:02,360 --> 00:00:06,680
Speaker 1: Bloomberg Audio Studios, podcasts, radio news.

2
00:00:06,920 --> 00:00:10,040
Speaker 2: Joining us now after extensive travel, thrilled you? Good, fine

3
00:00:10,080 --> 00:00:13,480
Speaker 2: time to start us off this morning, Daniel Day us

4
00:00:13,520 --> 00:00:15,720
Speaker 2: in Singapore. How do you get back to like the

5
00:00:15,760 --> 00:00:18,239
Speaker 2: golf stream always got the web bush golf stream blowblight

6
00:00:18,239 --> 00:00:21,279
Speaker 2: out of technology for web bushes, Stan I just I

7
00:00:21,320 --> 00:00:25,799
Speaker 2: want to look forward. What did you learn from Nvidia yesterday?

8
00:00:26,640 --> 00:00:30,600
Speaker 2: That's part of the AI story of the other ten stocks.

9
00:00:31,840 --> 00:00:34,479
Speaker 3: I mean it's all about Blackwell that's fueling. I mean

10
00:00:34,560 --> 00:00:38,720
Speaker 3: eleven billion that shows. I mean this is manage rejectory

11
00:00:38,800 --> 00:00:42,239
Speaker 3: fifteen to twenty billion a quarter. So if you had

12
00:00:42,360 --> 00:00:47,000
Speaker 3: any concerns, it's the opposite Big tech, other enterprises, they're

13
00:00:47,040 --> 00:00:51,120
Speaker 3: doubling down. That's the multiplier factor cross tech. For every

14
00:00:51,159 --> 00:00:54,000
Speaker 3: dollar spent on a video chip, eight ten dollars spent

15
00:00:54,080 --> 00:00:55,080
Speaker 3: across to us attack.

16
00:00:55,560 --> 00:00:58,160
Speaker 2: What is your ex exis do when you look at

17
00:00:58,600 --> 00:01:01,720
Speaker 2: technology develop and we'll talk to man deep sing about

18
00:01:01,720 --> 00:01:05,039
Speaker 2: this at a moment, dan Ie, do you have like

19
00:01:05,360 --> 00:01:08,720
Speaker 2: a timeline out on Nvidia? Do you have a terminal

20
00:01:08,840 --> 00:01:11,720
Speaker 2: value on in video that's appropriate seven years?

21
00:01:11,760 --> 00:01:12,200
Speaker 1: What is it?

22
00:01:13,360 --> 00:01:16,720
Speaker 3: I think two to two twenty five to me is

23
00:01:16,840 --> 00:01:18,840
Speaker 3: ultimately going to be sort of the base cases this

24
00:01:18,920 --> 00:01:21,399
Speaker 3: plays out, and again you're not gonna need like the

25
00:01:21,480 --> 00:01:25,640
Speaker 3: massive blowouts. They stay in this trajectory outside of some

26
00:01:25,760 --> 00:01:27,840
Speaker 3: black swan and the Bears will always hold for that

27
00:01:27,920 --> 00:01:30,679
Speaker 3: like they have the last decade. I think this is

28
00:01:30,720 --> 00:01:33,880
Speaker 3: a four an ultimately five trillion dollars markup as it

29
00:01:33,920 --> 00:01:37,800
Speaker 3: plays out. There is one red food in the world

30
00:01:38,000 --> 00:01:40,760
Speaker 3: for these chips, and he's wearing a black leather jacket.

31
00:01:40,840 --> 00:01:42,880
Speaker 3: His name is Jensen Dan.

32
00:01:42,959 --> 00:01:45,160
Speaker 4: Talk to us about the just catalysts for this name.

33
00:01:45,360 --> 00:01:48,520
Speaker 4: You know, it's it's really unusual for Nvidia shareholders to

34
00:01:48,520 --> 00:01:51,080
Speaker 4: look down at their Bloomberg screen and see year to

35
00:01:51,200 --> 00:01:53,640
Speaker 4: date sucks, you know, down a couple of percent here,

36
00:01:54,000 --> 00:01:55,600
Speaker 4: not to see green on the screen here. What's the

37
00:01:55,640 --> 00:01:58,240
Speaker 4: next catalyst do you think for this stock?

38
00:01:59,600 --> 00:02:04,800
Speaker 3: Yeah, I think catalysts now clearly going forward, you're going

39
00:02:04,880 --> 00:02:08,080
Speaker 3: to continue to focus on what Blackwell demand looks like.

40
00:02:08,240 --> 00:02:10,560
Speaker 3: Right and in the video you know, they'll have their

41
00:02:10,600 --> 00:02:13,320
Speaker 3: conference in March. I think that's going to be a

42
00:02:13,400 --> 00:02:16,280
Speaker 3: sort of spectacle that's gonna you know, they're launching pad

43
00:02:16,760 --> 00:02:20,440
Speaker 3: for Jensen Foreign video and look, and I think there

44
00:02:20,480 --> 00:02:23,160
Speaker 3: was all smoke, no fire, and I just continue. And

45
00:02:23,200 --> 00:02:27,600
Speaker 3: we've talked about on the show Deep Seek is bullish

46
00:02:28,040 --> 00:02:31,800
Speaker 3: for in video and big tech. It's not the opposite,

47
00:02:31,960 --> 00:02:34,079
Speaker 3: despite you know many of the barrizone fire in a

48
00:02:34,120 --> 00:02:34,639
Speaker 3: crowd theater.

49
00:02:34,919 --> 00:02:37,000
Speaker 2: If you're just joining us, folks at Man Deep sign

50
00:02:37,040 --> 00:02:39,200
Speaker 2: will be with us in a moment. From Bloomberg Intelligence,

51
00:02:39,200 --> 00:02:41,960
Speaker 2: we start with Daniel Ives of web Bush. They're head

52
00:02:42,000 --> 00:02:44,840
Speaker 2: of technology. For those of you on Bloomberg Radio, it

53
00:02:44,880 --> 00:02:48,600
Speaker 2: is critical to point out Ives with the Penn State

54
00:02:48,720 --> 00:02:54,359
Speaker 2: Knitney Lions merch in the background this morning at Penn's State,

55
00:02:54,520 --> 00:02:56,800
Speaker 2: Dan Eyes, they're going to have to do a rebuy

56
00:02:57,000 --> 00:03:00,959
Speaker 2: of Apple technology. We have Mark Germi in four or five,

57
00:03:01,080 --> 00:03:04,960
Speaker 2: six days ago and we started with a chip and

58
00:03:05,160 --> 00:03:08,160
Speaker 2: you know it's like Man Deep sing talk and Dan

59
00:03:08,200 --> 00:03:13,600
Speaker 2: and I's German was riveting about the technological veracity of

60
00:03:13,639 --> 00:03:16,000
Speaker 2: the Apple story. Do you agree?

61
00:03:17,320 --> 00:03:17,760
Speaker 1: Oh?

62
00:03:17,919 --> 00:03:22,160
Speaker 3: I think the best is yet to come because I did,

63
00:03:22,240 --> 00:03:25,080
Speaker 3: and German's talked about it as well. But the consumer

64
00:03:25,160 --> 00:03:29,600
Speaker 3: AI revolution comes from Coupertino. They're a toll collector. China

65
00:03:29,639 --> 00:03:33,120
Speaker 3: app us APP Australia doesn't matter. In other words, it's

66
00:03:33,160 --> 00:03:35,640
Speaker 3: about that. In stallwas I continue to think that is

67
00:03:35,680 --> 00:03:39,720
Speaker 3: going to incrementally add sixty seventy dollars per share the

68
00:03:39,760 --> 00:03:43,560
Speaker 3: AI story, the incremental services, and that doesn't include a

69
00:03:43,560 --> 00:03:46,040
Speaker 3: new form factor chip and just more and more control

70
00:03:46,080 --> 00:03:47,040
Speaker 3: of their ecosystem.

71
00:03:48,760 --> 00:03:52,000
Speaker 4: Dan, how did the company kind of position the deep

72
00:03:52,080 --> 00:03:56,280
Speaker 4: seek issue as a competitor out there in the marketplace?

73
00:03:57,440 --> 00:03:59,920
Speaker 3: I think actually someone of the rest of the big ten.

74
00:04:00,400 --> 00:04:04,560
Speaker 3: It's actually it's a positive the models. The more innovative

75
00:04:04,640 --> 00:04:08,119
Speaker 3: the models get, the cheaper they get, the more use case,

76
00:04:08,200 --> 00:04:11,400
Speaker 3: the more GPU capacity. In other words, like that's why

77
00:04:11,520 --> 00:04:14,360
Speaker 3: like that because everyone here and you guys talk about

78
00:04:14,400 --> 00:04:17,200
Speaker 3: all the time, like everyone's waiting for where's the cracks

79
00:04:17,200 --> 00:04:20,640
Speaker 3: in the armor smoke? Is this the AI bubble? But

80
00:04:20,680 --> 00:04:23,080
Speaker 3: we've talked about this two trillion of cap backs next

81
00:04:23,080 --> 00:04:25,919
Speaker 3: three years, Big Tech span three hundred and thirty billion

82
00:04:26,040 --> 00:04:29,599
Speaker 3: this year. You can find cracks, but you're not going

83
00:04:29,680 --> 00:04:31,840
Speaker 3: to find it here. And I think that's what you

84
00:04:31,880 --> 00:04:35,720
Speaker 3: saw last night in terms of really the demand there

85
00:04:35,800 --> 00:04:39,360
Speaker 3: was no I thought Gens was actually more bullish than

86
00:04:39,440 --> 00:04:41,240
Speaker 3: even what he was a few months ago.

87
00:04:41,520 --> 00:04:43,560
Speaker 2: Daniel, I thank you so much. Get some sleep too

88
00:04:43,600 --> 00:04:48,320
Speaker 2: early traveling worldwide with US with wet Bush well rested.

89
00:04:48,400 --> 00:04:51,840
Speaker 2: Mandeep Singh joins us right now, senior technology analysts for

90
00:04:51,880 --> 00:04:55,520
Speaker 2: Bloomberg Intelligence. The distinction here man deep designs is doing

91
00:04:55,600 --> 00:04:58,200
Speaker 2: by old cell trying to make money. You're looking at

92
00:04:58,200 --> 00:05:02,880
Speaker 2: the industry holistically. If I'm such in a della and

93
00:05:02,960 --> 00:05:08,680
Speaker 2: I'm listening in Vidia yesterday, what did Microsoft learn from

94
00:05:08,680 --> 00:05:10,920
Speaker 2: the the Nvidia conference call?

95
00:05:12,240 --> 00:05:15,760
Speaker 1: Yeah, look, I think Dan mentioned about black Vell ramping

96
00:05:15,839 --> 00:05:19,440
Speaker 1: up nicely. Contributed eleven billion dollars in the quarter, So

97
00:05:19,520 --> 00:05:23,159
Speaker 1: that means that video is really ramping up nicely with

98
00:05:23,279 --> 00:05:25,680
Speaker 1: its new architecture. In fact, they laid out the plan

99
00:05:25,800 --> 00:05:29,960
Speaker 1: for black Belt Ultra, so they're moving very fast when

100
00:05:29,960 --> 00:05:33,360
Speaker 1: it comes to the advances with the chips. So one

101
00:05:33,440 --> 00:05:36,400
Speaker 1: I would say, you know, data point that didn't do

102
00:05:36,520 --> 00:05:39,159
Speaker 1: very well was on the networking side, so they had

103
00:05:39,640 --> 00:05:43,400
Speaker 1: a second sequential decline in networking revenue. And if I

104
00:05:43,640 --> 00:05:46,040
Speaker 1: try to tie that in, you know with the size

105
00:05:46,080 --> 00:05:49,799
Speaker 1: of clusters. Again, in AI, there are these two main concepts,

106
00:05:49,839 --> 00:05:54,120
Speaker 1: training versus inferencing. Training is driven by the size of clusters,

107
00:05:54,160 --> 00:05:56,599
Speaker 1: and then if you have a bigger cluster, you have

108
00:05:56,760 --> 00:06:00,160
Speaker 1: to network it in a way where it's coher and

109
00:06:00,200 --> 00:06:03,680
Speaker 1: then you know you could use it for your training purposes.

110
00:06:03,760 --> 00:06:07,760
Speaker 1: So the fact that in Vidia's networking revenue decline sequentially,

111
00:06:07,800 --> 00:06:11,440
Speaker 1: again it's off a very strong base, it tells you

112
00:06:11,560 --> 00:06:14,000
Speaker 1: that there may be a slight pause when it comes

113
00:06:14,080 --> 00:06:18,719
Speaker 1: to building these bigger clusters and Nvidio is the most

114
00:06:18,760 --> 00:06:22,919
Speaker 1: exposed to that you know trend. So from that perspective,

115
00:06:23,000 --> 00:06:27,480
Speaker 1: I think you could ask yourself given their valuation, I mean,

116
00:06:27,520 --> 00:06:30,560
Speaker 1: clearly a lot is baked in in terms of growth

117
00:06:30,640 --> 00:06:34,080
Speaker 1: and the outer years is what everyone is worried about.

118
00:06:34,320 --> 00:06:37,680
Speaker 1: Can Nvidia sustain this momentum in you know, the outer

119
00:06:37,760 --> 00:06:41,320
Speaker 1: years twenty twenty seven, twenty eight and beyond. And I

120
00:06:41,320 --> 00:06:43,520
Speaker 1: don't think we got an answer last night in terms

121
00:06:43,600 --> 00:06:45,920
Speaker 1: of how this could drive it. And the other data

122
00:06:45,960 --> 00:06:48,600
Speaker 1: point I would call is the China sales. I mean

123
00:06:48,640 --> 00:06:51,960
Speaker 1: it's half of where it was before they export controls,

124
00:06:52,240 --> 00:06:55,400
Speaker 1: and their guide doesn't bake in the fact that if

125
00:06:55,520 --> 00:06:58,760
Speaker 1: tariffs go into effect then it would be a much

126
00:06:58,760 --> 00:07:01,800
Speaker 1: bigger hit on the roast margin, which actually fell three

127
00:07:01,839 --> 00:07:02,520
Speaker 1: hundred paces on.

128
00:07:03,480 --> 00:07:05,680
Speaker 4: So, following up on that, mat Man Deep, how did

129
00:07:05,720 --> 00:07:09,480
Speaker 4: Gensi Zwang kind of characterize the whle tariff environment, that

130
00:07:09,640 --> 00:07:13,880
Speaker 4: trade environment between China and the West.

131
00:07:15,120 --> 00:07:19,120
Speaker 1: Yeah, he didn't explain much. Neither did the CFO in

132
00:07:19,200 --> 00:07:22,680
Speaker 1: terms of what could go wrong. I mean, as of now,

133
00:07:22,760 --> 00:07:26,720
Speaker 1: what they said was China sales is half of before

134
00:07:26,760 --> 00:07:30,760
Speaker 1: the export controls went into effect under Biden. And look,

135
00:07:30,880 --> 00:07:33,520
Speaker 1: if they tightened the export controls, which is a very

136
00:07:33,680 --> 00:07:36,480
Speaker 1: likely scenario given what we have heard from the administration,

137
00:07:37,200 --> 00:07:40,400
Speaker 1: then China sales could go to zero by the end

138
00:07:40,440 --> 00:07:44,120
Speaker 1: of a year. And when you think about upside, I mean,

139
00:07:44,160 --> 00:07:46,800
Speaker 1: given Nvidia has said this trend of beat and races,

140
00:07:46,880 --> 00:07:49,280
Speaker 1: which they did last night as well, you have to

141
00:07:49,320 --> 00:07:52,080
Speaker 1: ask to yourself where will that upside come from. If

142
00:07:52,200 --> 00:07:54,600
Speaker 1: China sales, which was as big as you know, twenty

143
00:07:54,640 --> 00:07:57,480
Speaker 1: percent of the overall sales at one point, now it's

144
00:07:57,600 --> 00:08:01,640
Speaker 1: much smaller around ten percent. Will will that have an

145
00:08:01,680 --> 00:08:04,560
Speaker 1: impact in terms of you know, the BDA and a restaurant.

146
00:08:04,640 --> 00:08:07,280
Speaker 2: Joining us on your commuter crass nation. Mandeep Singh with

147
00:08:07,320 --> 00:08:11,160
Speaker 2: us with Bloomberg Intelligence on your technology holdings