1
00:00:02,520 --> 00:00:07,040
Speaker 1: Bloomberg Audio Studios, Podcasts, radio News.

2
00:00:10,960 --> 00:00:14,280
Speaker 2: Welcome to the Bloomberg Daybreak Asia podcast. I'm Deck Chrisner.

3
00:00:14,640 --> 00:00:17,600
Speaker 2: On today's episode, we'll be taking a deep dive into

4
00:00:17,680 --> 00:00:21,920
Speaker 2: Deep Seek. This is a Chinese artificial intelligence startup and

5
00:00:22,040 --> 00:00:25,360
Speaker 2: its aimodel, known as R one, appears to rival the

6
00:00:25,440 --> 00:00:30,400
Speaker 2: performance of US products like Chat GPT. Perhaps its biggest

7
00:00:30,440 --> 00:00:33,279
Speaker 2: selling point is that the model was developed at a

8
00:00:33,320 --> 00:00:36,440
Speaker 2: fraction of the cost of competing products, and that could

9
00:00:36,440 --> 00:00:40,640
Speaker 2: make high valuations of some USAI related companies a little

10
00:00:40,640 --> 00:00:43,640
Speaker 2: tough to justify. Let me give you one example today

11
00:00:44,000 --> 00:00:47,559
Speaker 2: in US trading, shares in the AI chip leader in Nvidia,

12
00:00:47,600 --> 00:00:50,680
Speaker 2: we're down nearly seventeen percent, and that led to a

13
00:00:50,680 --> 00:00:53,960
Speaker 2: loss in the Nasdaq one hundred of three percent. In

14
00:00:54,000 --> 00:00:56,320
Speaker 2: a moment, we'll be speaking with Jamie Cox. He is

15
00:00:56,360 --> 00:00:59,720
Speaker 2: managing partner at Harris Financial Group. But let's begin in

16
00:00:59,800 --> 00:01:03,960
Speaker 2: home with Bloomberg Tech editor of Vlad Savov. It's always

17
00:01:04,000 --> 00:01:06,240
Speaker 2: a pleasure to chat with you've Lad. This feels like

18
00:01:06,280 --> 00:01:10,520
Speaker 2: a major inflection point in the development of AI models. First,

19
00:01:10,560 --> 00:01:13,199
Speaker 2: let me get your reaction and what you're hearing among

20
00:01:13,319 --> 00:01:15,760
Speaker 2: Hong Kongers. Now that follow the tech space.

21
00:01:16,319 --> 00:01:19,240
Speaker 3: Well, the interesting thing about deep Seek is it's a

22
00:01:19,280 --> 00:01:24,319
Speaker 3: company that has rocketed to global renown and recognition today

23
00:01:24,440 --> 00:01:26,640
Speaker 3: and over the past week, but it has been known

24
00:01:26,680 --> 00:01:29,880
Speaker 3: about ever since it's beginning in twenty twenty three. It's

25
00:01:29,920 --> 00:01:33,120
Speaker 3: one that our Bloomberg Intelligence colleagues have been taken a

26
00:01:33,160 --> 00:01:36,759
Speaker 3: close look at because it is developing such a hot

27
00:01:36,840 --> 00:01:40,440
Speaker 3: area of new software as AI. It has been developing

28
00:01:40,440 --> 00:01:44,320
Speaker 3: its own models, so people here have been aware of it.

29
00:01:44,319 --> 00:01:48,240
Speaker 3: It has been on the radar. But really the thing

30
00:01:48,280 --> 00:01:52,200
Speaker 3: that has happened is it hit a certain threshold of recognition.

31
00:01:53,200 --> 00:01:57,080
Speaker 3: Part of it really was Silicon Valley people, venture capitalists.

32
00:01:57,160 --> 00:02:00,520
Speaker 3: They took an interest, they tried it out. They found

33
00:02:00,560 --> 00:02:03,600
Speaker 3: out that it is, indeed, like you mentioned, as competitive,

34
00:02:03,640 --> 00:02:07,760
Speaker 3: as the company itself says. It has posted some benchmark

35
00:02:07,840 --> 00:02:10,800
Speaker 3: results on this website. It has claimed already that it

36
00:02:10,840 --> 00:02:14,440
Speaker 3: competes with open Ai, but frankly, every AI model developer

37
00:02:14,480 --> 00:02:17,360
Speaker 3: claims that they compete or they can rival open Ai.

38
00:02:17,840 --> 00:02:20,320
Speaker 3: The amazing thing, and the really really impressive thing is

39
00:02:20,320 --> 00:02:22,600
Speaker 3: that the people who know this stuff, which I can't

40
00:02:22,600 --> 00:02:25,320
Speaker 3: claim that I'm an expert in it, they all recognize

41
00:02:25,320 --> 00:02:27,720
Speaker 3: it as a breakthrough so we need to recognize that too.

42
00:02:28,240 --> 00:02:31,280
Speaker 3: And it's also worth mentioning here that the company has

43
00:02:31,280 --> 00:02:33,320
Speaker 3: put out a research paper of its own and they've

44
00:02:33,320 --> 00:02:36,680
Speaker 3: said that the costs of training this model are less

45
00:02:36,680 --> 00:02:39,320
Speaker 3: than six million dollars. Now, that excludes a whole bunch

46
00:02:39,360 --> 00:02:42,520
Speaker 3: of costs, so we can't be exactly sure about how

47
00:02:42,600 --> 00:02:46,160
Speaker 3: much more affordable it is for this small company to

48
00:02:46,200 --> 00:02:49,840
Speaker 3: develop it, but it is very clearly a difference maker

49
00:02:50,000 --> 00:02:52,160
Speaker 3: when it comes to our assessment of how much it

50
00:02:52,200 --> 00:02:54,560
Speaker 3: costs to develop a world leading AI model.

51
00:02:54,639 --> 00:02:57,200
Speaker 2: Yeah, I think to be fair, our analyst here in

52
00:02:57,240 --> 00:02:59,639
Speaker 2: New York, Man Deep Saying, pointed out that our one

53
00:02:59,760 --> 00:03:04,520
Speaker 2: was to developed using output from open Ai, Anthropic and

54
00:03:04,639 --> 00:03:07,320
Speaker 2: LAMA to train the models. So it's almost as though

55
00:03:07,480 --> 00:03:10,240
Speaker 2: AI is training AI a little bit. And I think

56
00:03:10,320 --> 00:03:14,120
Speaker 2: that what may have precipitated or at least engendered a

57
00:03:14,160 --> 00:03:16,880
Speaker 2: lot of the selling that we saw in the US

58
00:03:16,880 --> 00:03:19,600
Speaker 2: in the last session was the rating on the Apple

59
00:03:19,639 --> 00:03:22,959
Speaker 2: app Store. I think R one rose to the top level.

60
00:03:23,720 --> 00:03:28,160
Speaker 2: Can we agree though, that the current Western approach to AI.

61
00:03:28,800 --> 00:03:31,920
Speaker 2: I'm talking here about the reliance on those high end chips,

62
00:03:32,440 --> 00:03:37,120
Speaker 2: extensive computing power, and vast amounts of electricity. Is that

63
00:03:37,200 --> 00:03:39,200
Speaker 2: being challenged to some extent.

64
00:03:39,320 --> 00:03:42,080
Speaker 3: Yes, we can agree that it's challenged. We cannot agree

65
00:03:42,120 --> 00:03:45,200
Speaker 3: that it's a foregone conclusion that it's going to be

66
00:03:45,240 --> 00:03:47,920
Speaker 3: going away. One thing to bear in mind here, which

67
00:03:47,920 --> 00:03:51,240
Speaker 3: I think is very important, is that Deepsek's model has

68
00:03:51,280 --> 00:03:55,080
Speaker 3: delivered an impressive chatbot. It only works with text. So

69
00:03:55,440 --> 00:03:58,000
Speaker 3: what we have is a breakthrough and an impressive achievement,

70
00:03:58,120 --> 00:04:01,480
Speaker 3: but it's within one medium. One of the things that

71
00:04:01,520 --> 00:04:03,640
Speaker 3: you have to bear in mind is that the more

72
00:04:03,680 --> 00:04:06,000
Speaker 3: advanced chips, the ones that are making a video is

73
00:04:06,000 --> 00:04:09,040
Speaker 3: still a multi trillion dollar company, even with its massive

74
00:04:09,040 --> 00:04:12,080
Speaker 3: share price drop. The thing that drives it is that

75
00:04:12,240 --> 00:04:14,160
Speaker 3: you want those chips in order to get to the

76
00:04:14,200 --> 00:04:17,640
Speaker 3: next stage of AI, specifically generative AI. So at the

77
00:04:17,640 --> 00:04:21,680
Speaker 3: moment we're talking about generating text, but the next evolution

78
00:04:21,839 --> 00:04:24,599
Speaker 3: is going to be generating images. Generating video is stuff

79
00:04:24,600 --> 00:04:29,120
Speaker 3: that open ai and other services have already made major

80
00:04:29,120 --> 00:04:32,360
Speaker 3: advances in, and that's where deep seek is probably going

81
00:04:32,400 --> 00:04:34,400
Speaker 3: to need a lot more hardware than it's used just

82
00:04:34,440 --> 00:04:34,880
Speaker 3: for text.

83
00:04:34,920 --> 00:04:37,680
Speaker 2: So it may be too much to say that deep

84
00:04:37,760 --> 00:04:41,040
Speaker 2: seek has been able to engineer a way around the

85
00:04:41,080 --> 00:04:44,360
Speaker 2: export bands that were put in place during the Biden administration,

86
00:04:45,000 --> 00:04:48,279
Speaker 2: the bands that basically kept China from accessing some of

87
00:04:48,320 --> 00:04:53,559
Speaker 2: those advanced semiconductors manufactured by and Vidiot. We can't really

88
00:04:53,760 --> 00:04:55,000
Speaker 2: articulate that, can we?

89
00:04:55,560 --> 00:04:58,240
Speaker 3: Yes, this is correct. Now, there are two things to

90
00:04:58,320 --> 00:05:03,279
Speaker 3: say here. Firstly, those bands and those trade cubs. Everyone

91
00:05:03,480 --> 00:05:06,000
Speaker 3: in the industry said, this is only going to encourage

92
00:05:06,120 --> 00:05:10,120
Speaker 3: China and Chinese companies to focus on more efficient methods.

93
00:05:10,320 --> 00:05:13,279
Speaker 3: So this is exactly what is being predicted by the industry.

94
00:05:13,520 --> 00:05:16,200
Speaker 3: They said, China will focus on more light, more small

95
00:05:16,240 --> 00:05:19,839
Speaker 3: scale models, and Deep Seek is the fruits of those efforts.

96
00:05:20,440 --> 00:05:23,120
Speaker 3: That being said, we cannot say that the sanctions are

97
00:05:23,120 --> 00:05:27,400
Speaker 3: not effective because Deep Seek CEO and founder has himself said,

98
00:05:27,800 --> 00:05:30,560
Speaker 3: the thing that Deep Seak requires right now is not

99
00:05:30,640 --> 00:05:34,680
Speaker 3: more money, is not more investment, is actually access to

100
00:05:34,800 --> 00:05:37,560
Speaker 3: those most advanced ships. The embargo is e cosed, the

101
00:05:37,640 --> 00:05:40,200
Speaker 3: ban on accessing those video chips is the thing that's

102
00:05:40,200 --> 00:05:41,200
Speaker 3: holding Deep Sea back.

103
00:05:41,480 --> 00:05:44,599
Speaker 2: Are we likely to see some type of fragmentation within

104
00:05:44,680 --> 00:05:48,039
Speaker 2: the artificial intelligence industry, which is to say that you

105
00:05:48,120 --> 00:05:51,719
Speaker 2: have certain products that have been produced at low cost

106
00:05:51,880 --> 00:05:56,320
Speaker 2: being rolled out at lower prices, and maybe those products

107
00:05:56,360 --> 00:05:59,440
Speaker 2: will see greater adoption and some of the more sophisticated

108
00:05:59,480 --> 00:06:02,919
Speaker 2: elements of artificial intelligence you'll have to pay a premium for,

109
00:06:03,600 --> 00:06:06,240
Speaker 2: and that segment of the market will be controlled by

110
00:06:07,000 --> 00:06:08,200
Speaker 2: very few companies.

111
00:06:08,640 --> 00:06:09,480
Speaker 4: Yes, I can very.

112
00:06:09,400 --> 00:06:11,440
Speaker 3: Much see that. One of the things that has been

113
00:06:11,560 --> 00:06:14,560
Speaker 3: the trend over the past few months is actually opening

114
00:06:14,640 --> 00:06:17,800
Speaker 3: Ey's revenue generation has accelerated, and it's one of the

115
00:06:17,839 --> 00:06:20,240
Speaker 3: trends that we're seeing. And if you had asked me

116
00:06:20,320 --> 00:06:22,239
Speaker 3: to predict what would be the big trend in AI

117
00:06:22,240 --> 00:06:25,560
Speaker 3: in twenty twenty five, it is exactly this taking all

118
00:06:25,600 --> 00:06:29,200
Speaker 3: that investment, taking all that developed technology, and turning into

119
00:06:29,240 --> 00:06:32,520
Speaker 3: products that you can sell subscriptions or in other ways

120
00:06:32,680 --> 00:06:35,440
Speaker 3: entices people to spend money on them. That is still

121
00:06:35,480 --> 00:06:38,040
Speaker 3: going to continue to be the case the premium products.

122
00:06:38,080 --> 00:06:42,400
Speaker 3: Open ai sells a two hundred dollars per month pro product.

123
00:06:42,960 --> 00:06:45,640
Speaker 3: There is an audience for that. But that being said,

124
00:06:46,160 --> 00:06:50,240
Speaker 3: the deep seek approach, it releases the software as an

125
00:06:50,240 --> 00:06:53,360
Speaker 3: open source software, so you as an individual can take

126
00:06:53,400 --> 00:06:55,520
Speaker 3: that and piggyback on top of it and decide that

127
00:06:55,560 --> 00:06:58,240
Speaker 3: you want to build your own specific chatbook. And a

128
00:06:58,240 --> 00:07:00,839
Speaker 3: good analogy that I have for this is when you

129
00:07:00,839 --> 00:07:03,919
Speaker 3: think about banking. You have the traditional banks, the ones

130
00:07:03,920 --> 00:07:06,800
Speaker 3: with branches and all those overheads and costs, and then

131
00:07:06,839 --> 00:07:09,359
Speaker 3: you have the digit banks, the ones that are mobile only,

132
00:07:09,440 --> 00:07:12,520
Speaker 3: app only, and because of that efficiency that those other

133
00:07:12,600 --> 00:07:16,280
Speaker 3: banks have, they can offer you better rates and better office.

134
00:07:16,040 --> 00:07:19,400
Speaker 2: So to go back to the hardware software issue, it

135
00:07:19,480 --> 00:07:23,920
Speaker 2: seems like Chinese AI engineers have found a way to

136
00:07:24,160 --> 00:07:28,080
Speaker 2: work with the software given the limitations of the hardware

137
00:07:28,120 --> 00:07:30,560
Speaker 2: that they have been forced to use because of those

138
00:07:31,040 --> 00:07:34,920
Speaker 2: US export controls. But I'm wondering whether on the semiconductor

139
00:07:34,920 --> 00:07:38,320
Speaker 2: industry side of things in China that we may see

140
00:07:38,360 --> 00:07:41,200
Speaker 2: some sort of breakthrough when it comes to advanced AI

141
00:07:41,320 --> 00:07:43,960
Speaker 2: chips in the near future. Is that a possibility.

142
00:07:44,280 --> 00:07:46,880
Speaker 3: Yes, There's two things to say here. Firstly, from what

143
00:07:46,920 --> 00:07:49,200
Speaker 3: little we know about Deep Seek, and bear in mind

144
00:07:49,320 --> 00:07:52,080
Speaker 3: it's a very very young startup. It came out of

145
00:07:52,080 --> 00:07:55,800
Speaker 3: an AI hedge fund AI driven hedge fund, so we

146
00:07:55,840 --> 00:07:58,760
Speaker 3: still don't know that much about it. But from interviews

147
00:07:58,760 --> 00:08:00,840
Speaker 3: with the founder, he has said that most of his

148
00:08:00,960 --> 00:08:05,520
Speaker 3: engineers are graduates from China's universities. They are entirely domestic

149
00:08:05,720 --> 00:08:10,000
Speaker 3: Chinese engineers, so credit to the company is educational system

150
00:08:10,360 --> 00:08:14,400
Speaker 3: now on the semiconductor front, there are technical limitations and challenges.

151
00:08:14,520 --> 00:08:18,360
Speaker 3: The classic one is EUV machines that only ASML makes

152
00:08:18,520 --> 00:08:22,560
Speaker 3: and is prevented from selling to China. Solving that issue.

153
00:08:22,600 --> 00:08:25,040
Speaker 3: We haven't seen much evidence of China doing it, but

154
00:08:25,080 --> 00:08:27,400
Speaker 3: then again, it may be something along the lines of

155
00:08:27,440 --> 00:08:29,720
Speaker 3: deep Seek. We never saw deep Seek. We were only

156
00:08:29,760 --> 00:08:31,920
Speaker 3: aware that is building a model. We never saw it

157
00:08:32,320 --> 00:08:35,319
Speaker 3: doing a breakthrough until today, until it was actually ready

158
00:08:35,360 --> 00:08:36,280
Speaker 3: to come out there.

159
00:08:36,480 --> 00:08:38,440
Speaker 2: Flight, We'll leave it there, Thank you so much. Let's

160
00:08:38,480 --> 00:08:41,520
Speaker 2: have of Bloomberg Tech editor in Asia joining us from

161
00:08:41,559 --> 00:08:51,439
Speaker 2: Hong Kong here on the Daybreak Asia podcast. Welcome back

162
00:08:51,440 --> 00:08:55,040
Speaker 2: to the Daybreak Asia Podcast. I'm Doug Chrisner. We continue

163
00:08:55,040 --> 00:08:58,480
Speaker 2: with our deep dive on deep Seek. News of the

164
00:08:58,559 --> 00:09:01,960
Speaker 2: Chinese startups new Chat bought r one, sparked to sell

165
00:09:01,960 --> 00:09:05,600
Speaker 2: off in many AI related names globally, and it wiped

166
00:09:05,640 --> 00:09:08,600
Speaker 2: out nearly a trillion dollars in market cap from US

167
00:09:08,600 --> 00:09:12,120
Speaker 2: and European tech names. So now a key question is

168
00:09:12,200 --> 00:09:15,920
Speaker 2: on capex spending plans for some of America's biggest companies.

169
00:09:16,160 --> 00:09:18,480
Speaker 2: Joining me now for a closer look. As Jamie Cox

170
00:09:19,000 --> 00:09:22,520
Speaker 2: is managing partner at Harris Financial Group. Jamie, thank you

171
00:09:22,559 --> 00:09:25,000
Speaker 2: so much for being with us. Certainly was an interesting

172
00:09:25,080 --> 00:09:28,360
Speaker 2: day where AI was concerned. How do you make sense

173
00:09:28,600 --> 00:09:31,520
Speaker 2: of the market reaction to this Deep Seek story.

174
00:09:31,880 --> 00:09:34,960
Speaker 4: I think the market's reaction was quite positive. I mean,

175
00:09:35,000 --> 00:09:35,959
Speaker 4: there were only.

176
00:09:35,800 --> 00:09:39,040
Speaker 1: Eight dollstocks that were down today. Sixty percent of the

177
00:09:39,320 --> 00:09:41,960
Speaker 1: S and P five hundred companies were up. So this

178
00:09:42,120 --> 00:09:46,800
Speaker 1: was very much an isolated day for you know, chip stocks.

179
00:09:46,840 --> 00:09:50,040
Speaker 1: This was sort of the moment in time where you

180
00:09:50,120 --> 00:09:53,959
Speaker 1: see disruption take hold, and when you start to see

181
00:09:53,960 --> 00:10:01,520
Speaker 1: things like AI disruption and the possibility that maybe companies

182
00:10:01,559 --> 00:10:05,600
Speaker 1: are overinvesting in AI infrastructure, you have to sit back

183
00:10:05,600 --> 00:10:08,160
Speaker 1: and think a minute and ask yourself, you know, if

184
00:10:08,280 --> 00:10:11,200
Speaker 1: Nvidia can legitimately sell thirty thousand dollars chips or not.

185
00:10:11,600 --> 00:10:14,280
Speaker 1: So I think it made a lot of investors think.

186
00:10:14,559 --> 00:10:17,679
Speaker 1: But a lot of investors have been worried about overvaluation in.

187
00:10:17,640 --> 00:10:18,559
Speaker 4: These stocks anyway.

188
00:10:18,600 --> 00:10:23,000
Speaker 1: So I think that the Deep Seek situation sort of precipitated.

189
00:10:23,480 --> 00:10:26,560
Speaker 1: Was the precipitating event for things that may have happened anyway.

190
00:10:26,720 --> 00:10:28,800
Speaker 1: So I think that's one thing to consider.

191
00:10:28,880 --> 00:10:31,280
Speaker 4: The second is I actually think this is a good thing.

192
00:10:32,040 --> 00:10:36,480
Speaker 1: It's very important that we not get COMPLACENTECH in our

193
00:10:36,520 --> 00:10:39,720
Speaker 1: technology prowess in the United States. We need to realize

194
00:10:39,720 --> 00:10:42,920
Speaker 1: that these technologies can be created at scale all over

195
00:10:42,920 --> 00:10:43,360
Speaker 1: the world.

196
00:10:44,000 --> 00:10:45,360
Speaker 4: And what Deep.

197
00:10:45,160 --> 00:10:47,880
Speaker 1: Seek basically did was kick off what I believe will

198
00:10:47,920 --> 00:10:51,880
Speaker 1: be an arms race, maybe a spot Nick moment for

199
00:10:51,960 --> 00:10:55,000
Speaker 1: the United States where we get really serious in the

200
00:10:55,720 --> 00:10:58,760
Speaker 1: prospect of competition from around the world will actually propel

201
00:10:59,240 --> 00:11:02,960
Speaker 1: AI developed forward. So I'm actually very positive both on

202
00:11:03,600 --> 00:11:08,840
Speaker 1: the development of AI and actually, and ironically, even though

203
00:11:08,840 --> 00:11:10,959
Speaker 1: the stocks we took a beating today, I actually think

204
00:11:11,000 --> 00:11:14,920
Speaker 1: that it further illustrates the need to build out AI

205
00:11:15,040 --> 00:11:19,000
Speaker 1: infrastructure because what we're seeing now is disruption in basically

206
00:11:19,080 --> 00:11:21,839
Speaker 1: AI one point zero, which will bring about the possibility

207
00:11:21,880 --> 00:11:25,000
Speaker 1: of AI two point zero, which will make it universally

208
00:11:25,080 --> 00:11:28,280
Speaker 1: usable and much more widely adopted than what we see

209
00:11:28,360 --> 00:11:31,120
Speaker 1: right now. So I'm pretty excited about what we see.

210
00:11:31,280 --> 00:11:33,160
Speaker 1: This is just a moment in time in the market,

211
00:11:33,640 --> 00:11:35,840
Speaker 1: and we'll be looking back on this as one of

212
00:11:35,840 --> 00:11:38,680
Speaker 1: the best buying opportunities in these stocks at quite some time.

213
00:11:38,960 --> 00:11:41,480
Speaker 2: So one of the things that I've found very compelling

214
00:11:41,520 --> 00:11:44,760
Speaker 2: here is that deep Seek's chatbot, known as R one

215
00:11:45,840 --> 00:11:50,400
Speaker 2: was apparently trained using less than cutting edge semiconductors, and

216
00:11:50,480 --> 00:11:55,480
Speaker 2: that in turn required much less energy consumption. So I

217
00:11:55,520 --> 00:11:57,280
Speaker 2: hear what you're saying when it comes to the sell

218
00:11:57,320 --> 00:12:00,960
Speaker 2: off that we had in certain semiconductor names today. But

219
00:12:01,080 --> 00:12:04,640
Speaker 2: I was also struck by the fact that Constellation Energy

220
00:12:04,720 --> 00:12:08,520
Speaker 2: was down twenty percent. How does the Deep Seek story

221
00:12:08,679 --> 00:12:10,640
Speaker 2: change the conversation around power?

222
00:12:11,559 --> 00:12:14,200
Speaker 4: So there's two parts of that question. I'll address the

223
00:12:14,200 --> 00:12:16,319
Speaker 4: power one first. There have been.

224
00:12:17,720 --> 00:12:23,520
Speaker 1: Reports of inability for companies to get adequate power sources

225
00:12:23,559 --> 00:12:29,160
Speaker 1: to data centers. So companies like Microsoft Amazon are putting

226
00:12:29,160 --> 00:12:34,080
Speaker 1: in personal nuclear reactors to actually power data centers, and

227
00:12:34,160 --> 00:12:37,800
Speaker 1: companies like Constellation Energy provide those services. So if the

228
00:12:37,920 --> 00:12:39,880
Speaker 1: data centers are not going to be as plentible and

229
00:12:39,920 --> 00:12:43,520
Speaker 1: therefore the power sources will not be needed, that's why

230
00:12:43,520 --> 00:12:45,199
Speaker 1: you see these knock on effects. It's one of the

231
00:12:45,280 --> 00:12:49,319
Speaker 1: reasons why the power story had been one of the

232
00:12:49,679 --> 00:12:53,320
Speaker 1: sort of tertiary ways to play the AI story without

233
00:12:53,360 --> 00:12:56,120
Speaker 1: actually buying the tech stocks. So that's why they're sort

234
00:12:56,160 --> 00:12:59,600
Speaker 1: of connected together in a trade.

235
00:12:59,880 --> 00:13:03,120
Speaker 4: I don't believe that.

236
00:13:02,320 --> 00:13:05,160
Speaker 1: Either story will turn out to be, you know, the

237
00:13:05,200 --> 00:13:06,760
Speaker 1: sell off today will be the right one.

238
00:13:06,800 --> 00:13:08,600
Speaker 4: I think it's actually the wrong choice.

239
00:13:08,960 --> 00:13:12,480
Speaker 1: But to go back to talk about the clever technology

240
00:13:12,559 --> 00:13:17,360
Speaker 1: or the clever engineering that made this possible, where the

241
00:13:17,600 --> 00:13:21,000
Speaker 1: R one, you know, learning module is able to use.

242
00:13:22,440 --> 00:13:25,200
Speaker 4: Just a just a It's a very small change, but

243
00:13:25,240 --> 00:13:26,520
Speaker 4: it's very very impactful.

244
00:13:26,559 --> 00:13:29,840
Speaker 1: So if you think about like disruptive technologies, you have

245
00:13:30,000 --> 00:13:34,840
Speaker 1: incumbent players who basically iterate they have built this fantastic

246
00:13:35,440 --> 00:13:38,440
Speaker 1: black box, and they just iterate small and small pieces,

247
00:13:38,480 --> 00:13:41,560
Speaker 1: Whereas these disruptive players come come around and they actually

248
00:13:41,600 --> 00:13:44,480
Speaker 1: change the entire game. And that's what Deep Seek has done.

249
00:13:44,679 --> 00:13:47,720
Speaker 1: They basically took the AI engine which is ever present

250
00:13:47,800 --> 00:13:52,720
Speaker 1: in open ai, where the technology is ready for any

251
00:13:52,800 --> 00:13:56,920
Speaker 1: question on any topic, whereas in the in the deep

252
00:13:56,920 --> 00:14:01,240
Speaker 1: Seek infrastructure, it would basically take the input and only

253
00:14:01,320 --> 00:14:04,040
Speaker 1: turn on the pieces of the infrastructure that were needed

254
00:14:04,080 --> 00:14:04,800
Speaker 1: to answer.

255
00:14:04,480 --> 00:14:05,600
Speaker 4: The question or to iterate.

256
00:14:05,840 --> 00:14:09,600
Speaker 1: So it basically didn't have the entire system running, you know,

257
00:14:09,679 --> 00:14:13,440
Speaker 1: all the time. And that will be readily adopted by

258
00:14:13,520 --> 00:14:18,960
Speaker 1: all AI providers. For sure, Beta and for sure Open

259
00:14:19,000 --> 00:14:22,080
Speaker 1: AI will adopt that clever engineering and they too will

260
00:14:22,120 --> 00:14:24,160
Speaker 1: have it. And so the reason why I think that

261
00:14:24,240 --> 00:14:27,479
Speaker 1: this is just the beginning is that now that infrastructure

262
00:14:27,840 --> 00:14:32,280
Speaker 1: that was being built to handle an ever present AI

263
00:14:32,440 --> 00:14:36,280
Speaker 1: infrastructure now could be used for other things. And I

264
00:14:36,280 --> 00:14:39,720
Speaker 1: think that's why we're going to see this AI infrastructure

265
00:14:39,760 --> 00:14:43,160
Speaker 1: actually blossom as a result of this disruption, not be

266
00:14:43,280 --> 00:14:43,880
Speaker 1: torn apart.

267
00:14:44,040 --> 00:14:46,480
Speaker 2: It is a critical week for tech earnings. This week.

268
00:14:46,520 --> 00:14:50,800
Speaker 2: Four of the mag seven are reporting. That's Apple, Microsoft, Tesla,

269
00:14:51,040 --> 00:14:54,160
Speaker 2: and Meta Platforms. What are your expectations here and what

270
00:14:54,200 --> 00:14:57,560
Speaker 2: do these companies need to say about the future in

271
00:14:57,680 --> 00:15:01,360
Speaker 2: order for investors to really kind of not lose heart.

272
00:15:02,120 --> 00:15:04,560
Speaker 4: Well, two of the four of them had a very

273
00:15:04,560 --> 00:15:05,240
Speaker 4: good day to day.

274
00:15:05,280 --> 00:15:07,960
Speaker 1: I mean, Apple and Meta really bucked the trend, and

275
00:15:08,040 --> 00:15:11,760
Speaker 1: it was largely because they're consumers of these more expensive,

276
00:15:12,360 --> 00:15:16,000
Speaker 1: you know, chips, and I think if anything that lowers

277
00:15:16,000 --> 00:15:20,280
Speaker 1: the infrastructure costs a cruise directly to players like Meta

278
00:15:20,560 --> 00:15:24,520
Speaker 1: or Apple. Apple's unique story because there have been some

279
00:15:25,120 --> 00:15:29,720
Speaker 1: concerns about the inability to produce or sell the iPhone.

280
00:15:30,200 --> 00:15:32,960
Speaker 1: Maybe sales had declined, maybe people weren't upgrading at the

281
00:15:33,000 --> 00:15:36,720
Speaker 1: same rate. But the story on Apple is less hardware

282
00:15:36,800 --> 00:15:39,840
Speaker 1: these days and definitely more about.

283
00:15:39,200 --> 00:15:40,680
Speaker 4: You know, the ancillary services.

284
00:15:40,680 --> 00:15:42,400
Speaker 1: So I think Apple is going to do quite well

285
00:15:42,440 --> 00:15:45,440
Speaker 1: and I think the stock definitely had a fantastic day

286
00:15:45,480 --> 00:15:47,640
Speaker 1: to day on the prospects of them, you know, having

287
00:15:47,720 --> 00:15:50,800
Speaker 1: higher margins because they're spending less on their data infrastructure

288
00:15:51,080 --> 00:15:54,360
Speaker 1: meta the same thing. I think Microsoft kind of got

289
00:15:54,400 --> 00:15:57,080
Speaker 1: caught up in the AI given its investment in the

290
00:15:57,160 --> 00:16:01,280
Speaker 1: in the open AI platform. But I think that Microsoft

291
00:16:01,320 --> 00:16:04,800
Speaker 1: is actually, of the three, the best position to continue

292
00:16:04,800 --> 00:16:07,640
Speaker 1: to grow earnings year over year, and so I think

293
00:16:07,680 --> 00:16:09,920
Speaker 1: three of the four are doing well. I'm not I

294
00:16:09,920 --> 00:16:13,680
Speaker 1: don't follow Tesla that much, but I feel like Elon

295
00:16:13,800 --> 00:16:16,360
Speaker 1: Musk may be in a better position to understand whether

296
00:16:16,440 --> 00:16:20,680
Speaker 1: or not this deep seek technology is actually you know,

297
00:16:20,680 --> 00:16:23,320
Speaker 1: whether or not the technology being deployed or the things

298
00:16:23,360 --> 00:16:26,000
Speaker 1: that the company has said that they've been able to do,

299
00:16:26,120 --> 00:16:29,520
Speaker 1: or actual reality because Tesla has one of the largest

300
00:16:29,520 --> 00:16:33,680
Speaker 1: AI infrastructures powered by all the automobiles that Tesla makes,

301
00:16:33,880 --> 00:16:37,440
Speaker 1: so their neural net has been around for some time

302
00:16:37,520 --> 00:16:41,160
Speaker 1: and has been generating AI content for generation of software

303
00:16:41,240 --> 00:16:43,800
Speaker 1: with the cars. So I think maybe we might learn

304
00:16:43,840 --> 00:16:47,240
Speaker 1: a thing or two in the conference call that may

305
00:16:47,360 --> 00:16:50,200
Speaker 1: be related to deep seek in some of this AI disruption.

306
00:16:50,280 --> 00:16:52,720
Speaker 1: I think that will be the story for the for

307
00:16:52,880 --> 00:16:54,320
Speaker 1: the Tesla conference call this.

308
00:16:54,320 --> 00:16:57,160
Speaker 2: Year, Jamie. Before I let you go, We've also have

309
00:16:57,200 --> 00:17:00,920
Speaker 2: a FED decision on Wednesday, Powell news conference right after

310
00:17:00,960 --> 00:17:04,240
Speaker 2: we get the decision. Is it likely that we're going

311
00:17:04,280 --> 00:17:06,600
Speaker 2: to get kind of a hawkish message right now?

312
00:17:07,160 --> 00:17:10,680
Speaker 4: No, I think the FED is on hold.

313
00:17:11,400 --> 00:17:15,160
Speaker 1: They may discuss, as they did in the press conference

314
00:17:15,200 --> 00:17:19,399
Speaker 1: and December, the need to be cautious about lowering rates

315
00:17:19,480 --> 00:17:24,119
Speaker 1: too fast. But it's going to be interesting to watch

316
00:17:24,119 --> 00:17:26,320
Speaker 1: over the next couple of months because the President is

317
00:17:26,359 --> 00:17:28,720
Speaker 1: going to put enormous pressure on the FED to do

318
00:17:28,760 --> 00:17:32,960
Speaker 1: something about mortgage rates. That has been a particular concern

319
00:17:33,280 --> 00:17:36,200
Speaker 1: to President Trump, and he is going to start beating

320
00:17:36,200 --> 00:17:38,919
Speaker 1: the drum on that very loudly. So I have a

321
00:17:38,920 --> 00:17:41,120
Speaker 1: feeling we'll start to see the FED start to talk

322
00:17:41,200 --> 00:17:43,040
Speaker 1: about how they're going to deal with the balance sheet

323
00:17:43,560 --> 00:17:44,800
Speaker 1: and how they could.

324
00:17:44,680 --> 00:17:47,120
Speaker 4: Help consumers on the mortgage side.

325
00:17:47,840 --> 00:17:49,960
Speaker 1: So I don't think this FED meeting will be much

326
00:17:49,960 --> 00:17:52,480
Speaker 1: on the FED fund rate in terms of news. I mean,

327
00:17:52,480 --> 00:17:55,080
Speaker 1: the FED fund futures basically are ninety nine percent that

328
00:17:55,160 --> 00:17:59,840
Speaker 1: they hold. What is not built into the model is

329
00:18:00,040 --> 00:18:02,440
Speaker 1: what they talk about on the ballot sheet, and there

330
00:18:02,480 --> 00:18:06,439
Speaker 1: has been runoff at a fairly decent equipped over the

331
00:18:06,480 --> 00:18:08,480
Speaker 1: last year and a half, and I think that may

332
00:18:08,520 --> 00:18:11,600
Speaker 1: be ready to slow down. And if the FED would

333
00:18:12,080 --> 00:18:15,040
Speaker 1: you slow its mortgage backed securities runoff, I think that

334
00:18:15,119 --> 00:18:17,200
Speaker 1: might help mortgage rates some. So I think that's where

335
00:18:17,200 --> 00:18:19,119
Speaker 1: you're going to see some of the action. And it

336
00:18:19,200 --> 00:18:22,280
Speaker 1: would not surprise me at all if the President is out,

337
00:18:22,440 --> 00:18:25,440
Speaker 1: you know, with comments in and around the FED meeting,

338
00:18:25,440 --> 00:18:28,000
Speaker 1: that would you know, really encourage them to be more

339
00:18:28,160 --> 00:18:30,960
Speaker 1: aggressive in helping mortgage rates to.

340
00:18:30,920 --> 00:18:33,719
Speaker 2: Con Jamie will leave it there. Thank you so much

341
00:18:33,760 --> 00:18:36,720
Speaker 2: for being with us. Jimmy Cox there, managing partner at

342
00:18:36,760 --> 00:18:40,280
Speaker 2: Harris Financial Group, joining from Richmond, Virginia. Here on the

343
00:18:40,359 --> 00:18:46,000
Speaker 2: Daybreak Asia Podcast. Thanks for listening to today's episode of

344
00:18:46,040 --> 00:18:50,160
Speaker 2: the Bloomberg Daybreak Asia Edition podcast. Each weekday, we look

345
00:18:50,160 --> 00:18:53,960
Speaker 2: at the story shaping markets, finance, and geopolitics in the

346
00:18:54,000 --> 00:18:57,240
Speaker 2: Asia Pacific. You can find us on Apple, Spotify, the

347
00:18:57,280 --> 00:19:01,280
Speaker 2: Bloomberg Podcast YouTube channel, or anywhere else you listen. Join

348
00:19:01,400 --> 00:19:04,399
Speaker 2: us again tomorrow for insight on the market moves from

349
00:19:04,400 --> 00:19:08,879
Speaker 2: Hong Kong to Singapore and Australia. I'm Doug Prisoner and

350
00:19:09,000 --> 00:19:10,160
Speaker 2: this is Bloomberg