1
00:00:02,480 --> 00:00:13,240
Speaker 1: Bloomberg Audio Studios, Podcasts, Radio News. Welcome to the Daybreak

2
00:00:13,280 --> 00:00:16,920
Speaker 1: Asia podcast. I'm Doug Krisner. We begin with the simmering

3
00:00:17,000 --> 00:00:20,400
Speaker 1: tensions between China and Japan, the trace back to late

4
00:00:20,480 --> 00:00:24,520
Speaker 1: last year with comments from Japanese Prime Minister Sana atuki

5
00:00:24,520 --> 00:00:28,120
Speaker 1: Ichi regarding Taiwan. For a closer look, I'm joined by

6
00:00:28,160 --> 00:00:32,479
Speaker 1: Bloomberg's Garfield Reynolds. He is team leader for Asia Markets

7
00:00:32,479 --> 00:00:36,400
Speaker 1: live on the Bloomberg terminal. Garfield joins from Sydney. Thank

8
00:00:36,440 --> 00:00:38,680
Speaker 1: you so much for being with us. Happy New Year

9
00:00:38,720 --> 00:00:41,600
Speaker 1: to you. Can you begin by giving me a little

10
00:00:41,600 --> 00:00:46,120
Speaker 1: bit of understanding the context of how we got here

11
00:00:46,200 --> 00:00:50,320
Speaker 1: with relation to the tension between China and Japan.

12
00:00:51,360 --> 00:00:55,200
Speaker 2: Yeah, Well, how we got here is that Japanese Prime

13
00:00:55,200 --> 00:01:00,600
Speaker 2: Minister Sana Takaichi, soon after she won her way to

14
00:01:00,680 --> 00:01:07,160
Speaker 2: offers through a parliamentary vote, commented about Taiwan and the

15
00:01:07,240 --> 00:01:12,160
Speaker 2: idea that Japan might be willing to play a role

16
00:01:12,600 --> 00:01:18,440
Speaker 2: in as matters developed there, and that brought on an

17
00:01:18,480 --> 00:01:24,839
Speaker 2: immediate response from China. You're recalling that, of course China.

18
00:01:25,120 --> 00:01:30,559
Speaker 2: China's official stance is that Taiwan is part of China

19
00:01:30,760 --> 00:01:33,840
Speaker 2: and will always be that it's not in any ways

20
00:01:33,880 --> 00:01:36,920
Speaker 2: a separate entity. It has no sovereignty, and of course

21
00:01:36,959 --> 00:01:40,840
Speaker 2: these days there are very few countries that recognize Taiwan

22
00:01:40,959 --> 00:01:44,960
Speaker 2: is a sovereign nation. So that's how we got here,

23
00:01:45,760 --> 00:01:50,240
Speaker 2: and to some extent, the heat came out of the issue.

24
00:01:50,600 --> 00:01:55,760
Speaker 2: But now China's come out with this again, and this

25
00:01:55,960 --> 00:02:00,560
Speaker 2: ramps up what's been going on, and that that's part

26
00:02:00,560 --> 00:02:02,640
Speaker 2: of the reason, a big part of the reason, I

27
00:02:02,680 --> 00:02:06,560
Speaker 2: would argue, why we're seeing Japanese shares quoted lower today,

28
00:02:07,080 --> 00:02:12,359
Speaker 2: even as Asian shares elsewhere, in particular South Korea continue

29
00:02:12,440 --> 00:02:16,760
Speaker 2: what has been a really stunning start through the year

30
00:02:17,040 --> 00:02:18,799
Speaker 2: for equities here in the region.

31
00:02:19,160 --> 00:02:22,200
Speaker 1: I was struck by the fact that Beijing recently banned

32
00:02:22,200 --> 00:02:26,520
Speaker 1: exports of what they are calling dual use goods to

33
00:02:26,600 --> 00:02:30,240
Speaker 1: the Japanese military. So there seems to be implicit in

34
00:02:30,280 --> 00:02:32,320
Speaker 1: all of this the real risk of some type of

35
00:02:32,639 --> 00:02:35,440
Speaker 1: confrontation or is it merely saber rattling do you think?

36
00:02:37,919 --> 00:02:41,679
Speaker 2: I think it's more saber rattling, and very much Beijing

37
00:02:43,919 --> 00:02:45,919
Speaker 2: seems to be acting on the basis that it doesn't

38
00:02:46,080 --> 00:02:51,880
Speaker 2: need to sort of directly physically as it were, threatened Japan,

39
00:02:52,480 --> 00:02:57,000
Speaker 2: rather it can hit Japan through its economy. Last year,

40
00:02:57,680 --> 00:03:04,639
Speaker 2: it discourages Chinese tourists from going to Japan, and that

41
00:03:04,800 --> 00:03:08,959
Speaker 2: was definitely something that could do some damage to Japan's economy.

42
00:03:09,560 --> 00:03:14,480
Speaker 2: We saw a number of tourism and also retail related

43
00:03:14,639 --> 00:03:19,400
Speaker 2: equities in Japan underperform, precisely because of that.

44
00:03:19,480 --> 00:03:22,280
Speaker 1: You mentioned a moment Ago Garfield, the positivity that we're

45
00:03:22,320 --> 00:03:25,120
Speaker 1: seeing in the South Korean equity market, I would be

46
00:03:25,200 --> 00:03:27,560
Speaker 1: remiss if we didn't kind of come back to the

47
00:03:27,600 --> 00:03:32,520
Speaker 1: meeting between Chinese President Chi Jinping and South Korean leader

48
00:03:32,720 --> 00:03:37,600
Speaker 1: Lee j Jung in Beijing. Helped me understand the importance

49
00:03:37,640 --> 00:03:40,080
Speaker 1: of this meeting and how it may be reflected in

50
00:03:40,120 --> 00:03:42,360
Speaker 1: some of the price action that we're seeing in markets.

51
00:03:43,160 --> 00:03:48,200
Speaker 2: South Korea's new president, relatively newly minted president, took over

52
00:03:48,320 --> 00:03:52,600
Speaker 2: after the previous president was ousted over what was seen

53
00:03:52,640 --> 00:03:57,280
Speaker 2: as some authoritarian moves there. So the new president's been

54
00:03:57,280 --> 00:04:02,240
Speaker 2: doing a pretty decent job of navigating some difficult waters

55
00:04:02,320 --> 00:04:05,600
Speaker 2: between the US on one side and China on the other.

56
00:04:06,240 --> 00:04:08,680
Speaker 2: On the one hand, the US is a lot term

57
00:04:08,720 --> 00:04:14,920
Speaker 2: ally of South Korea, and yet President Trump had imposed tariffs,

58
00:04:15,440 --> 00:04:18,919
Speaker 2: complained about the trade surplus, so Korea mostly seemed to

59
00:04:18,960 --> 00:04:22,280
Speaker 2: have gotten him on side. Meantime, they've also been working

60
00:04:22,320 --> 00:04:25,560
Speaker 2: hard to get China back on side. There was a

61
00:04:25,680 --> 00:04:29,839
Speaker 2: Chinese I think there still is a Chinese ban on

62
00:04:30,240 --> 00:04:36,400
Speaker 2: some cultural exports as it were from Korea movies, videos, music,

63
00:04:36,920 --> 00:04:40,320
Speaker 2: which doesn't necessarily do a lot of economic damage, but

64
00:04:40,880 --> 00:04:46,040
Speaker 2: is not a good look from the Korean's point of view.

65
00:04:46,120 --> 00:04:50,400
Speaker 2: So they're very eager to work with China to get

66
00:04:50,440 --> 00:04:55,120
Speaker 2: them on side along with the US. And although there's

67
00:04:55,160 --> 00:04:59,799
Speaker 2: no sort of direct idea that Korean shares are rallying

68
00:05:00,000 --> 00:05:04,440
Speaker 2: because of china diplomatic efforts, there is an argument to

69
00:05:04,480 --> 00:05:07,799
Speaker 2: be made that part of what's helping South Korean equities

70
00:05:07,800 --> 00:05:12,480
Speaker 2: outperform is that they are seen as being a strong

71
00:05:13,160 --> 00:05:19,440
Speaker 2: AI slash tech play that is not necessarily tied to

72
00:05:19,640 --> 00:05:26,560
Speaker 2: one or other of the US AI slash LLM streams.

73
00:05:26,839 --> 00:05:31,680
Speaker 2: So in Nvidia versus Google, as it were, architecture, there,

74
00:05:32,200 --> 00:05:36,039
Speaker 2: South Korean companies plug into both. They also have some

75
00:05:36,160 --> 00:05:40,159
Speaker 2: exposure to China's AI boom and what's going on with

76
00:05:40,240 --> 00:05:45,560
Speaker 2: deep Seek, so we're seeing the costs be dramatically outperformed

77
00:05:45,680 --> 00:05:51,600
Speaker 2: US equities as the AI related boom for global stocks extends.

78
00:05:51,960 --> 00:05:54,080
Speaker 1: I want to change gears. Garfield, talk a little bit

79
00:05:54,120 --> 00:05:57,840
Speaker 1: about the oil market today. WTI crude oil right now

80
00:05:57,920 --> 00:06:00,800
Speaker 1: is down more than one point eight percent, and in

81
00:06:00,839 --> 00:06:05,360
Speaker 1: the US Tuesday evening, President Trump said that interim authorities

82
00:06:05,400 --> 00:06:09,760
Speaker 1: in Venezuela will be turning over between thirty to fifty

83
00:06:09,839 --> 00:06:14,800
Speaker 1: million barrels of high qualitied sanctioned oil to the United States. Now,

84
00:06:14,839 --> 00:06:17,359
Speaker 1: we know that China is a big probably the biggest

85
00:06:17,480 --> 00:06:20,560
Speaker 1: consumer of Venezuelan and crude. Talk to me a little

86
00:06:20,560 --> 00:06:24,000
Speaker 1: bit about what the reaction has been to the developments

87
00:06:24,000 --> 00:06:27,240
Speaker 1: that we have seen over the last several days in Venezuela,

88
00:06:27,279 --> 00:06:30,960
Speaker 1: and whether or not the US involvement now in the

89
00:06:30,960 --> 00:06:35,760
Speaker 1: crude oil industry in Venezuela could be a significant issue

90
00:06:35,800 --> 00:06:39,840
Speaker 1: for the Chinese.

91
00:06:38,160 --> 00:06:42,400
Speaker 2: Impact for the crude markets of what's gone in Venezuela.

92
00:06:42,800 --> 00:06:46,440
Speaker 2: What's gone on in Venezuela should be a lower pricing

93
00:06:46,520 --> 00:06:52,240
Speaker 2: crude and a lack of any scope for serious rallies,

94
00:06:52,440 --> 00:06:55,919
Speaker 2: sustained rallies because whatever may or may not go on

95
00:06:56,040 --> 00:06:59,960
Speaker 2: and reviving Venezuela's oil industry would take plenty of time

96
00:07:00,120 --> 00:07:03,760
Speaker 2: and money. The simple fact that you have a bias

97
00:07:03,800 --> 00:07:07,359
Speaker 2: towards more supplies coming out of Venezuela than the trickle

98
00:07:07,720 --> 00:07:11,800
Speaker 2: that has been coming recently. That adds to the potential

99
00:07:11,800 --> 00:07:17,040
Speaker 2: for a supply glut, So lower crude prices globally is

100
00:07:17,120 --> 00:07:19,400
Speaker 2: part of what you would expect with the way the

101
00:07:19,480 --> 00:07:22,600
Speaker 2: Venezuelan situation has been developing. From the point of view

102
00:07:22,640 --> 00:07:25,560
Speaker 2: of what that means for China, on the one hand,

103
00:07:25,880 --> 00:07:29,240
Speaker 2: it has been the biggest buyer of Venezuelan crude and

104
00:07:29,400 --> 00:07:33,760
Speaker 2: it's Venezuela's biggest creditor, so it's a little surprise that

105
00:07:33,880 --> 00:07:38,240
Speaker 2: China was very critical of the US moves in Venezuela.

106
00:07:38,360 --> 00:07:43,840
Speaker 2: But it is noticeable that there hasn't been any really

107
00:07:44,200 --> 00:07:49,640
Speaker 2: heated rhetoric or moves out of China. There's been no

108
00:07:49,680 --> 00:07:53,520
Speaker 2: sense that, for example, they might impose breash restrictions on

109
00:07:53,640 --> 00:07:56,240
Speaker 2: exports to the US, such as those they've just toweded

110
00:07:56,240 --> 00:07:59,920
Speaker 2: for Japan. So that signal was a couple of things.

111
00:08:00,000 --> 00:08:02,160
Speaker 2: I think one is that, of course China can get

112
00:08:02,280 --> 00:08:08,520
Speaker 2: crued from elsewhere, and if global crude prices are falling,

113
00:08:09,200 --> 00:08:12,920
Speaker 2: that means that the cost for China to get that

114
00:08:13,080 --> 00:08:17,880
Speaker 2: crude is not going to be significantly up, and China

115
00:08:17,920 --> 00:08:22,200
Speaker 2: had also been stockpiling some crude, so I think the

116
00:08:22,280 --> 00:08:26,640
Speaker 2: hit to China might not be too huge. China also,

117
00:08:26,680 --> 00:08:32,559
Speaker 2: of course, has famously boosted its usage of renewables, precisely

118
00:08:32,800 --> 00:08:37,480
Speaker 2: because that's a way of increasing its energy independence. So

119
00:08:37,840 --> 00:08:41,000
Speaker 2: there is a potential modest hit to China from what

120
00:08:41,040 --> 00:08:42,960
Speaker 2: the US is doing on the as far as ven

121
00:08:43,000 --> 00:08:45,760
Speaker 2: as well as crude goes, but as I said, probably

122
00:08:45,880 --> 00:08:49,640
Speaker 2: just a modest one. Of course, it also has no

123
00:08:49,720 --> 00:08:52,600
Speaker 2: problems getting crued out of Russia, so that could be

124
00:08:52,600 --> 00:08:55,880
Speaker 2: a help. And there's also from the point of view

125
00:08:55,920 --> 00:08:59,760
Speaker 2: of China coming back to where we started with regards

126
00:08:59,760 --> 00:09:05,679
Speaker 2: to Taiwan. China does want very much to ultimately absorb Taiwan.

127
00:09:06,240 --> 00:09:08,240
Speaker 2: And if it looks at what the US has done

128
00:09:08,800 --> 00:09:13,040
Speaker 2: in saying, well, the western hemisphere, Monroe doctrine says that

129
00:09:13,120 --> 00:09:16,040
Speaker 2: the western hemisphere is the US's backyard, and we get

130
00:09:16,080 --> 00:09:20,120
Speaker 2: to say what happens here all China that turns around

131
00:09:20,160 --> 00:09:23,679
Speaker 2: and says, we agree with this sort of thing in fact,

132
00:09:24,080 --> 00:09:26,120
Speaker 2: or you know, you certainly can't tell us to lay

133
00:09:26,160 --> 00:09:33,640
Speaker 2: off Taiwan when you're busy, you know, running roughshod over Venezuela.

134
00:09:34,040 --> 00:09:36,760
Speaker 2: I wouldn't expand it affect China to do anything as

135
00:09:36,840 --> 00:09:40,360
Speaker 2: dramatic as what's been occurring the last couple of weeks

136
00:09:40,440 --> 00:09:47,040
Speaker 2: or so with Venezuela, but it does enhance China's capacity

137
00:09:47,120 --> 00:09:49,600
Speaker 2: to put pressure on Taiwan and to see a path

138
00:09:49,640 --> 00:09:56,000
Speaker 2: toward at some stage absorbing it. So any pain from

139
00:09:56,040 --> 00:10:00,880
Speaker 2: the Venezuelan crude side of things, it might be well

140
00:10:00,880 --> 00:10:03,360
Speaker 2: and truly okay with China when it looks at the

141
00:10:03,360 --> 00:10:05,760
Speaker 2: longer term as it usually does and say, well, this

142
00:10:05,920 --> 00:10:09,040
Speaker 2: opens up a path for us to have access at

143
00:10:09,040 --> 00:10:14,760
Speaker 2: some stage to the extremely valuable Taiwanese economy, and we.

144
00:10:14,760 --> 00:10:19,320
Speaker 1: Know that Chinese President Chi Jinping has set unification with

145
00:10:19,559 --> 00:10:23,200
Speaker 1: Taiwan as one of his goals. Garfield, thank you so much.

146
00:10:23,320 --> 00:10:27,440
Speaker 1: Bloomberg's Garfield Reynolds, team leader for Asia Markets, Live on

147
00:10:27,440 --> 00:10:30,559
Speaker 1: the Bloomberg Terminal. Garfield joining from Sydney here on the

148
00:10:30,640 --> 00:10:41,160
Speaker 1: Daybreak Asia podcast. Welcome back to the Daybreak Asia Podcast.

149
00:10:41,200 --> 00:10:44,360
Speaker 1: I'm Doug Krisner. We go to Las Vegas Next and

150
00:10:44,640 --> 00:10:48,520
Speaker 1: cees the annual event focused on consumer tech. That's where

151
00:10:48,520 --> 00:10:52,000
Speaker 1: we caught up with Nvidia CEO Jensen Huang today he

152
00:10:52,160 --> 00:10:55,640
Speaker 1: joined the CEO of Siemens, Roland Bush to discuss the

153
00:10:55,679 --> 00:11:00,000
Speaker 1: partnership of these two companies. They spoke with Bloomberg's at Ludlow.

154
00:11:00,120 --> 00:11:03,880
Speaker 3: And we've discussed the five layer cake often. You know

155
00:11:03,960 --> 00:11:08,360
Speaker 3: within video. It started with the GPUs, but it's now software,

156
00:11:08,400 --> 00:11:12,000
Speaker 3: the element of simulation. On stage, you talked about integrating

157
00:11:12,400 --> 00:11:15,800
Speaker 3: the software side, in particular into DA Again, a similar

158
00:11:15,840 --> 00:11:17,880
Speaker 3: question for you, more of a timeline. What is it

159
00:11:18,000 --> 00:11:20,040
Speaker 3: you think you'll do first in how you guys work

160
00:11:20,080 --> 00:11:20,679
Speaker 3: closer together.

161
00:11:20,840 --> 00:11:23,560
Speaker 4: The first thing, let me just say very quickly, we're

162
00:11:23,600 --> 00:11:26,440
Speaker 4: announcing a big partnership between us. We've known each other

163
00:11:26,520 --> 00:11:29,280
Speaker 4: for a long time, but the partnership we're announcing is

164
00:11:29,320 --> 00:11:30,040
Speaker 4: really a big deal.

165
00:11:30,800 --> 00:11:31,240
Speaker 5: One.

166
00:11:31,400 --> 00:11:37,120
Speaker 4: We're accelerating their EDA software. We're accelerating their simulation software.

167
00:11:37,520 --> 00:11:42,880
Speaker 4: We're integrating AI technology, physical AI and agentic AI into

168
00:11:43,000 --> 00:11:47,600
Speaker 4: their team center and their factory automation operating system, and

169
00:11:47,679 --> 00:11:51,480
Speaker 4: so we're working together across this entire spectrum. When we

170
00:11:51,559 --> 00:11:54,040
Speaker 4: accelerate the software, then we'll get to use it to

171
00:11:54,080 --> 00:11:58,160
Speaker 4: design our chips and systems. When we accelerate their simulation software,

172
00:11:58,320 --> 00:12:01,120
Speaker 4: we'll use it in our AI factories to simulate the

173
00:12:01,120 --> 00:12:07,320
Speaker 4: thermal properties of our AI factories. When we integrate our

174
00:12:07,360 --> 00:12:12,960
Speaker 4: automation and agentic systems into their AI industrial operating system,

175
00:12:13,280 --> 00:12:16,000
Speaker 4: we can then use it in our factory floors with.

176
00:12:16,000 --> 00:12:17,720
Speaker 5: Our partners, for example fox Con.

177
00:12:18,160 --> 00:12:21,319
Speaker 4: And so we're working across this entire spectrum together and

178
00:12:21,360 --> 00:12:24,079
Speaker 4: we're going to put the technology to use basically as

179
00:12:24,080 --> 00:12:24,800
Speaker 4: soon as we can.

180
00:12:24,880 --> 00:12:28,199
Speaker 3: What's the net effect for you, Jensen? Is it improves

181
00:12:28,240 --> 00:12:31,320
Speaker 3: margins efficient capital allocation. I know that might sound a

182
00:12:31,320 --> 00:12:33,880
Speaker 3: bit dry, but actually, right now, that's the answer everyone's

183
00:12:33,880 --> 00:12:37,559
Speaker 3: searching for. How is this investment in AI and use

184
00:12:37,600 --> 00:12:40,160
Speaker 3: of the technology actually change things in the real world?

185
00:12:40,160 --> 00:12:42,440
Speaker 5: For me, now's yesterday Vera Rubin.

186
00:12:42,640 --> 00:12:46,959
Speaker 4: It takes six different chips to integrate into this incredible

187
00:12:47,000 --> 00:12:50,839
Speaker 4: system called Vera Rubin. And when you're done, each one

188
00:12:50,840 --> 00:12:55,000
Speaker 4: of these Vera Rubin GPUs is two hundred and forty

189
00:12:55,320 --> 00:12:59,679
Speaker 4: thousand wants and it is ten times more energy efficient

190
00:12:59,679 --> 00:13:03,120
Speaker 4: than all generation. It is ten times more cost efficient

191
00:13:03,160 --> 00:13:07,200
Speaker 4: than a last generation. But still the technology is insanely complicated.

192
00:13:07,440 --> 00:13:12,000
Speaker 4: One hundred fifteen thousand engineering years came together to build

193
00:13:12,000 --> 00:13:15,360
Speaker 4: this system. And so when we accelerate EDA tools, when

194
00:13:15,400 --> 00:13:19,560
Speaker 4: we accelerate stimulation tools and when we can eventually and

195
00:13:19,600 --> 00:13:24,119
Speaker 4: I'm hoping very soon design entire vera Ruben systems inside

196
00:13:24,120 --> 00:13:28,480
Speaker 4: a Semen's digital twin, the chance the ability for us

197
00:13:28,520 --> 00:13:32,680
Speaker 4: to create much much more complex systems will scale, will

198
00:13:32,720 --> 00:13:35,240
Speaker 4: do it much more efficiently, and so this is really

199
00:13:35,280 --> 00:13:38,160
Speaker 4: about being able to do the impossible and being able

200
00:13:38,200 --> 00:13:40,840
Speaker 4: to do it the impossible right.

201
00:13:40,880 --> 00:13:47,439
Speaker 5: The first time. And once they then realize that AI

202
00:13:47,679 --> 00:13:48,600
Speaker 5: creates real.

203
00:13:48,400 --> 00:13:52,480
Speaker 6: World impact, this is where it really deploys the full power.

204
00:13:52,320 --> 00:13:53,840
Speaker 5: And also the economic power.

205
00:13:54,400 --> 00:13:56,600
Speaker 6: It's not only in the data centers on any efactories

206
00:13:56,600 --> 00:13:58,720
Speaker 6: which we see, but also on the edge because once

207
00:13:58,760 --> 00:14:01,920
Speaker 6: we start influencing law latency, you bring this technology to

208
00:14:02,000 --> 00:14:05,040
Speaker 6: the edge. This is a huge potential for our customers

209
00:14:05,080 --> 00:14:07,360
Speaker 6: too to deploy this technology.

210
00:14:07,480 --> 00:14:10,719
Speaker 5: And this includes of course includes hardware where we come

211
00:14:10,760 --> 00:14:11,680
Speaker 5: from from the jibs.

212
00:14:12,400 --> 00:14:15,160
Speaker 6: It goes into the controllers, some of our controllers run

213
00:14:15,240 --> 00:14:18,160
Speaker 6: on GPUs and then it goes all the.

214
00:14:18,120 --> 00:14:24,000
Speaker 5: Way to the industrial PC exactly and are AI industry.

215
00:14:24,400 --> 00:14:28,640
Speaker 6: We supercharge it and they can now run algorithms trained

216
00:14:28,640 --> 00:14:29,200
Speaker 6: in the cloud.

217
00:14:29,240 --> 00:14:31,400
Speaker 5: They can run it on the shop floor and do

218
00:14:31,560 --> 00:14:34,200
Speaker 5: all that trick. What we talked about it.

219
00:14:34,200 --> 00:14:37,560
Speaker 6: In real time optimization and running in a plant, and

220
00:14:37,600 --> 00:14:38,600
Speaker 6: that makes a huge difference.

221
00:14:38,720 --> 00:14:40,120
Speaker 5: It drives the economy growth.

222
00:14:40,240 --> 00:14:42,280
Speaker 3: The question that's being searched for is how is this

223
00:14:42,360 --> 00:14:43,880
Speaker 3: going to manifest in the real world. You know, the

224
00:14:43,960 --> 00:14:46,920
Speaker 3: emphasis that this is cs I think is physical AI

225
00:14:47,600 --> 00:14:50,680
Speaker 3: is not the manifestation of the final stage of physical

226
00:14:50,720 --> 00:14:52,960
Speaker 3: AI just one giant robot that you guys call a

227
00:14:52,960 --> 00:14:56,480
Speaker 3: factory In the manufacturing context, is that where you're seeing

228
00:14:56,520 --> 00:14:59,240
Speaker 3: demand from actual customers, you know that they need a

229
00:14:59,280 --> 00:15:03,720
Speaker 3: factory that is automated and genuinely autonomous in some degree.

230
00:15:04,080 --> 00:15:06,480
Speaker 6: Number one is if you want to build a factory,

231
00:15:06,640 --> 00:15:10,360
Speaker 6: and very often you're missing out on labor. And we

232
00:15:10,520 --> 00:15:12,960
Speaker 6: talk about the own skilled labor, it's hard to find

233
00:15:13,080 --> 00:15:16,720
Speaker 6: number one. Number two is once you build it autonomous

234
00:15:16,760 --> 00:15:20,080
Speaker 6: and automated, then you have a much much higher yield,

235
00:15:20,200 --> 00:15:23,040
Speaker 6: but you can generate to use lesser energy, by the way,

236
00:15:23,040 --> 00:15:25,440
Speaker 6: at the same time before you optimize it in a way.

237
00:15:25,720 --> 00:15:27,440
Speaker 5: So therefore there's a lot of benefits.

238
00:15:28,000 --> 00:15:30,160
Speaker 6: And if you want to talk about the United States

239
00:15:30,840 --> 00:15:34,480
Speaker 6: up manufacturing United States, you need to go as digital

240
00:15:34,720 --> 00:15:38,000
Speaker 6: and as automated, and AI is supercharged as possible.

241
00:15:38,680 --> 00:15:44,880
Speaker 7: A factory is robotic, and it is orchestrating robots that

242
00:15:44,960 --> 00:15:48,600
Speaker 7: are building systems that are also robotic, like for example,

243
00:15:48,600 --> 00:15:50,120
Speaker 7: self dragging cars a robotic system.

244
00:15:50,840 --> 00:15:54,320
Speaker 4: And the reason why it's so hard to deploy robots

245
00:15:54,360 --> 00:15:57,840
Speaker 4: today is because it's hard to program these robotic systems.

246
00:15:58,560 --> 00:16:03,000
Speaker 4: The software expertise next necessary to customization necessary is really intense.

247
00:16:03,040 --> 00:16:06,240
Speaker 4: It's just too much. And so the fact that we

248
00:16:06,280 --> 00:16:11,280
Speaker 4: could now apply artificial intelligence physical AI technology to these

249
00:16:11,360 --> 00:16:16,240
Speaker 4: robotic systems make them easier to teach. You show it

250
00:16:16,320 --> 00:16:19,160
Speaker 4: a few demonstrations, and the AI learned it by itself.

251
00:16:19,520 --> 00:16:21,200
Speaker 3: I think, would you, guys far As say Jensen, that

252
00:16:21,240 --> 00:16:24,000
Speaker 3: you've solved for that that software limitation.

253
00:16:24,040 --> 00:16:27,480
Speaker 4: That's offtware limitation is the chat GPT moment of it

254
00:16:27,560 --> 00:16:30,040
Speaker 4: is now here. I think over the course of the

255
00:16:30,040 --> 00:16:31,920
Speaker 4: next couple of two three years, we're going to make

256
00:16:32,000 --> 00:16:33,240
Speaker 4: some really big breakthroughs.

257
00:16:33,760 --> 00:16:37,040
Speaker 3: Let's please talk about energy, electricity, power supply, call it

258
00:16:37,080 --> 00:16:39,960
Speaker 3: what you will. In turn, for each of you, how

259
00:16:40,000 --> 00:16:42,600
Speaker 3: worried are you about it as a bottleneck and what

260
00:16:42,760 --> 00:16:45,120
Speaker 3: is your experience date staying running both companies.

261
00:16:45,520 --> 00:16:48,360
Speaker 4: In that respect, you should always be a bottleneck for

262
00:16:48,400 --> 00:16:51,640
Speaker 4: any industry, and this is a new industry that's growing

263
00:16:51,760 --> 00:16:55,720
Speaker 4: incredibly fast. As you know, AI is both the technology

264
00:16:55,720 --> 00:16:58,360
Speaker 4: that's going to revolutionize many applications, and we're talking about

265
00:16:58,400 --> 00:17:03,400
Speaker 4: some of them here. The AI industry itself, the manufacturing

266
00:17:03,440 --> 00:17:06,840
Speaker 4: of the artificial intelligence takes energy. It takes energy, it

267
00:17:06,880 --> 00:17:10,920
Speaker 4: takes AI factories. It's exactly the reason why from Hopper

268
00:17:11,000 --> 00:17:16,200
Speaker 4: to Blackwell we increased energy efficiency by ten x. From

269
00:17:16,240 --> 00:17:20,879
Speaker 4: Blackwell to Reuben, we increased energy efficiency again by tenx.

270
00:17:21,560 --> 00:17:25,840
Speaker 4: And that translates directly to our customers' revenues because in

271
00:17:25,880 --> 00:17:31,080
Speaker 4: the case of an AI factory, whatever factory size you have,

272
00:17:31,160 --> 00:17:34,359
Speaker 4: you're limited by the power and that power within that

273
00:17:34,400 --> 00:17:37,440
Speaker 4: power constraint, you want to have the most tokens or

274
00:17:37,480 --> 00:17:41,600
Speaker 4: most AI per whte that you can possibly generate. And

275
00:17:41,640 --> 00:17:45,760
Speaker 4: so every time we improve energy efficiency, we're effectively improving

276
00:17:46,320 --> 00:17:50,159
Speaker 4: both the AI capabilities for our customers and their revenues

277
00:17:50,680 --> 00:17:52,439
Speaker 4: because they're always constrained by power.

278
00:17:52,600 --> 00:17:56,440
Speaker 1: That was in Nvidia CEO Jensen Huang and Siemens CEO

279
00:17:56,640 --> 00:18:01,119
Speaker 1: Roland Bush speaking with Bloomberg zed Ludlow at CEES and

280
00:18:01,160 --> 00:18:04,920
Speaker 1: turned to artificial intelligence in China, we had the chance

281
00:18:04,960 --> 00:18:07,520
Speaker 1: to visit with Lisa Sue, the CEO of a m D.

282
00:18:08,160 --> 00:18:11,800
Speaker 1: She told us demand for AI in Beijing is high.

283
00:18:11,920 --> 00:18:14,360
Speaker 1: Here is Sue speaking with Bloomberg's at Ludlow.

284
00:18:14,680 --> 00:18:16,240
Speaker 8: China is an important market for us.

285
00:18:16,680 --> 00:18:18,600
Speaker 9: You know, We actually sell a broad range of chips

286
00:18:18,600 --> 00:18:22,040
Speaker 9: into China, including our you know, our PCs as well

287
00:18:22,080 --> 00:18:24,600
Speaker 9: as you know, other embedded chips as.

288
00:18:24,520 --> 00:18:27,600
Speaker 5: In the data sets context, sorry, in the data.

289
00:18:27,440 --> 00:18:30,520
Speaker 9: Center context, we are you know, certainly we see China

290
00:18:30,560 --> 00:18:31,720
Speaker 9: as an important market.

291
00:18:32,160 --> 00:18:34,760
Speaker 8: We were We did get some licenses.

292
00:18:34,240 --> 00:18:37,240
Speaker 9: From the US government, you know, late last year as

293
00:18:37,280 --> 00:18:40,119
Speaker 9: it relates to some of our previous generation our m

294
00:18:40,160 --> 00:18:43,000
Speaker 9: I three oh eight you know chips, and we are

295
00:18:43,040 --> 00:18:46,360
Speaker 9: in the process of applying for new licenses with our

296
00:18:46,880 --> 00:18:48,800
Speaker 9: m I three twenty five chips that.

297
00:18:48,760 --> 00:18:54,040
Speaker 8: Were recently allowed to apply for licenses.

298
00:18:54,040 --> 00:18:57,080
Speaker 9: We haven't gotten those licenses yet, but we continue to

299
00:18:57,160 --> 00:18:58,800
Speaker 9: view China as an important market for us.

300
00:18:58,920 --> 00:19:00,560
Speaker 3: The reason I asked about it is is in part

301
00:19:00,640 --> 00:19:03,199
Speaker 3: because a lot of the work that's being done in

302
00:19:03,240 --> 00:19:06,239
Speaker 3: open source models and bridging the gap between open and

303
00:19:06,280 --> 00:19:08,800
Speaker 3: closed it is being done in China. To some extent,

304
00:19:09,880 --> 00:19:11,800
Speaker 3: There's been a lot of discussion about the demands being

305
00:19:11,840 --> 00:19:14,120
Speaker 3: there in China, But could you reflect a little bit

306
00:19:14,119 --> 00:19:17,520
Speaker 3: on that demand, but also what the Chinese government's attitude

307
00:19:17,600 --> 00:19:22,800
Speaker 3: is to you taking a later generation of tech to

308
00:19:22,840 --> 00:19:23,320
Speaker 3: the country.

309
00:19:23,680 --> 00:19:26,840
Speaker 8: Well, I do think the demand for AI in.

310
00:19:26,760 --> 00:19:30,000
Speaker 9: General and in China is high for all the reasons

311
00:19:30,000 --> 00:19:32,200
Speaker 9: that we talked about. I think we are in a

312
00:19:32,240 --> 00:19:36,879
Speaker 9: demand environment where more compute is beneficial across the world.

313
00:19:37,320 --> 00:19:39,880
Speaker 9: We think China is an important market for US, and

314
00:19:39,960 --> 00:19:43,120
Speaker 9: it's a place where we certainly would have been very

315
00:19:43,160 --> 00:19:45,879
Speaker 9: active in having our solutions deployed, so, you know, we

316
00:19:45,960 --> 00:19:48,800
Speaker 9: continue to view it as something that's important. We're working

317
00:19:48,800 --> 00:19:51,399
Speaker 9: with the US government as well as our Chinese customers,

318
00:19:51,920 --> 00:19:53,840
Speaker 9: you know, to find good solutions there.

319
00:19:53,720 --> 00:19:56,320
Speaker 3: And there are signs from both governments that the licensed

320
00:19:56,359 --> 00:19:59,320
Speaker 3: process is moving. Commerce is kind of notorious for things

321
00:19:59,320 --> 00:20:00,960
Speaker 3: sitting on a desk for quite a long time.

322
00:20:01,040 --> 00:20:04,600
Speaker 9: I think we are optimistic that we'll have an opportunity

323
00:20:04,760 --> 00:20:06,400
Speaker 9: to get some of those licenses.

324
00:20:06,440 --> 00:20:10,280
Speaker 1: Granted, that was AMD CEO Lisa Sue speaking with Bloomberg's

325
00:20:10,359 --> 00:20:14,359
Speaker 1: Ed Ludlow at cees here on the Daybreak Asia podcast.

326
00:20:17,720 --> 00:20:21,080
Speaker 1: Thanks for listening to today's episode of the Bloomberg Daybreak

327
00:20:21,240 --> 00:20:24,600
Speaker 1: Asia Edition podcast. Each weekday, we look at the story

328
00:20:24,680 --> 00:20:29,040
Speaker 1: shaping markets, finance, and geopolitics in the Asia Pacific. You

329
00:20:29,080 --> 00:20:33,159
Speaker 1: can find us on Apple, Spotify, the Bloomberg Podcast YouTube channel,

330
00:20:33,280 --> 00:20:36,320
Speaker 1: or anywhere else you listen. Join us again tomorrow for

331
00:20:36,440 --> 00:20:39,919
Speaker 1: insight on the market moves from Hong Kong to Singapore

332
00:20:40,320 --> 00:20:44,080
Speaker 1: and Australia. I'm Doug Chrisner, and this is Bloomberg