WEBVTT - Nvidia's Upbeat Forecast, A Conversation with CEO Jensen Huang

0:00:00.040 --> 0:00:10.680
<v Speaker 1>Bloomberg Audio Studios, Podcasts, Radio News.

0:00:12.680 --> 0:00:15.800
<v Speaker 2>Welcome to the Daybreak Asia podcast. I'm Doug Krisner. We

0:00:15.920 --> 0:00:19.040
<v Speaker 2>begin with the results from the world's most valuable publicly

0:00:19.079 --> 0:00:22.440
<v Speaker 2>traded company after the bell in. Vidio posted third quarter

0:00:22.520 --> 0:00:25.919
<v Speaker 2>numbers above estimates, and the company gave a strong forecast

0:00:25.960 --> 0:00:29.000
<v Speaker 2>for revenue in the current period. Sales are expected to

0:00:29.040 --> 0:00:32.879
<v Speaker 2>be about sixty five billion dollars and gross margins are

0:00:32.920 --> 0:00:36.360
<v Speaker 2>expected to reach the mid seventy percent range. So this

0:00:36.560 --> 0:00:42.120
<v Speaker 2>outlook signals demand does remain strong for Invidia's artificial intelligence accelerators.

0:00:42.520 --> 0:00:45.479
<v Speaker 2>We took a deeper dive into the Nvidia story with

0:00:45.600 --> 0:00:49.080
<v Speaker 2>CEO Jensen Wong. He spoke with Bloomberg's Ed Ludlow.

0:00:49.200 --> 0:00:53.640
<v Speaker 3>Jensen, It's been an astonishingly busy day for you in Washington, DC,

0:00:53.760 --> 0:00:58.320
<v Speaker 3>so I'm grateful for your time. You're sold out of Blackwell,

0:00:58.720 --> 0:01:01.560
<v Speaker 3>that's what you said, and also that five hundred billion

0:01:01.560 --> 0:01:05.720
<v Speaker 3>dollar forecast, which is Blackwell Rubin has room to grow.

0:01:06.360 --> 0:01:07.640
<v Speaker 3>How do those fit together?

0:01:11.520 --> 0:01:18.160
<v Speaker 4>I said, Sales are off the charts for Blackwell and

0:01:18.280 --> 0:01:22.360
<v Speaker 4>Nvidia GPUs in the cloud are sold out. We got

0:01:22.400 --> 0:01:25.360
<v Speaker 4>plenty of Blackwells to sell you. We have lots of

0:01:25.400 --> 0:01:29.399
<v Speaker 4>Blackwells coming. We're making a lot of Blackwells, and we

0:01:29.480 --> 0:01:34.520
<v Speaker 4>have a bunch of Vera Rubens coming, and so business

0:01:34.600 --> 0:01:38.760
<v Speaker 4>is very very strong. But we've planned our supply chain

0:01:38.800 --> 0:01:41.520
<v Speaker 4>incredibly well. We have the largest supply chain in the world.

0:01:42.200 --> 0:01:47.840
<v Speaker 4>Our partners TSMC, our memory partners, sk Heinex, Micron, Samsung

0:01:48.040 --> 0:01:51.040
<v Speaker 4>are doing a fantastic job supporting us. And all of

0:01:51.120 --> 0:01:54.880
<v Speaker 4>our systems partners, fox Con and Quantum and Wistron are

0:01:54.920 --> 0:01:58.240
<v Speaker 4>packaging partners. Everybody's doing a fantastic job supporting us. And

0:01:58.280 --> 0:02:01.480
<v Speaker 4>we've done a good job planning for a very very

0:02:01.520 --> 0:02:03.880
<v Speaker 4>strong year, and we've done a good job planning for

0:02:03.960 --> 0:02:08.680
<v Speaker 4>Vera Rubin. So sales are off the charts and video

0:02:08.720 --> 0:02:11.760
<v Speaker 4>GPUs and the cloud is sold out, but we've got

0:02:11.760 --> 0:02:13.080
<v Speaker 4>a bunch of Blackwells to sell.

0:02:14.280 --> 0:02:17.000
<v Speaker 3>Jensen, what's the road ahead for Vera Rubin. It's one

0:02:17.000 --> 0:02:19.080
<v Speaker 3>of the most common questions we get for you of

0:02:19.120 --> 0:02:22.200
<v Speaker 3>how that ramp will go relative to what we saw

0:02:22.280 --> 0:02:25.400
<v Speaker 3>with the Blackwell generations.

0:02:27.120 --> 0:02:32.120
<v Speaker 4>Well, the silicon for Vera Rubin, seven different chips are

0:02:32.160 --> 0:02:34.800
<v Speaker 4>back in our labs and the bring up is happening

0:02:34.840 --> 0:02:38.920
<v Speaker 4>across engineering teams. Probably a couple of twenty thousand people

0:02:39.000 --> 0:02:43.240
<v Speaker 4>are working on bringing up Vera Rubin from silicon to systems,

0:02:43.280 --> 0:02:47.400
<v Speaker 4>the software to algorithms. People are working around the clock

0:02:47.880 --> 0:02:51.840
<v Speaker 4>and this bring up is going beautifully. We're on track

0:02:52.000 --> 0:02:56.240
<v Speaker 4>to deliver Vera Rubin about Q three timeframe of next year,

0:02:57.440 --> 0:03:02.040
<v Speaker 4>continuing our once a year cycle. Vera Ruben is already

0:03:03.040 --> 0:03:07.760
<v Speaker 4>assured a huge success. Everybody's incredibly excited about it. Can't

0:03:07.800 --> 0:03:11.760
<v Speaker 4>wait to show everybody. And then one last thing is

0:03:11.800 --> 0:03:16.760
<v Speaker 4>that the rack architecture, the rack scale architecture is completely revolutionary.

0:03:17.400 --> 0:03:21.120
<v Speaker 4>It includes a scale up switch called the mvy Link

0:03:21.760 --> 0:03:25.000
<v Speaker 4>envy Link seventy two. Our fifth generation is the only

0:03:25.040 --> 0:03:28.440
<v Speaker 4>one of its kind in the world. This rack architecture,

0:03:28.840 --> 0:03:32.960
<v Speaker 4>which is incredibly complex, started with Grace Blackwell, then Grace

0:03:33.000 --> 0:03:37.800
<v Speaker 4>Blackwell Ultra. It is transitioned to Grace Blackwell Ultra is

0:03:37.800 --> 0:03:43.760
<v Speaker 4>incredibly seamless. The same rack scale architecture is going to

0:03:43.800 --> 0:03:46.800
<v Speaker 4>be used for Vera Rubin, and so the supply chain

0:03:46.880 --> 0:03:50.600
<v Speaker 4>is all used to it. This complexity that we enjoyed

0:03:50.720 --> 0:03:55.240
<v Speaker 4>with Grace Blackwell transition, we're now incredibly smooth running and

0:03:55.280 --> 0:03:57.200
<v Speaker 4>so I think Vera Ruben is going to be just

0:03:57.440 --> 0:03:59.440
<v Speaker 4>really smooth and we're gonna rap it really hard.

0:04:00.440 --> 0:04:03.560
<v Speaker 3>Jensen I tried to go through what the CFO COLLECT

0:04:03.600 --> 0:04:07.200
<v Speaker 3>CRES said about China. In the quarter gone it seemed

0:04:07.240 --> 0:04:10.600
<v Speaker 3>like there was not meaning for H twenty sales because

0:04:10.640 --> 0:04:13.400
<v Speaker 3>the demand wasn't there even if you were permitted to

0:04:13.400 --> 0:04:16.040
<v Speaker 3>sell H twenty. And then in the current period and

0:04:16.080 --> 0:04:19.200
<v Speaker 3>going forward in video, seems committed to working with both

0:04:19.240 --> 0:04:24.640
<v Speaker 3>the United States and China to sell what COLLECT called

0:04:24.839 --> 0:04:28.480
<v Speaker 3>more competitive compute. Where do we stand with that and

0:04:28.839 --> 0:04:32.120
<v Speaker 3>could you just clarify what COLLECT was talking about in

0:04:32.160 --> 0:04:36.000
<v Speaker 3>the current state of play for China.

0:04:37.040 --> 0:04:40.719
<v Speaker 4>The most important thing she said is that we've said

0:04:40.760 --> 0:04:44.719
<v Speaker 4>for some time now our forecast for China is zero.

0:04:45.800 --> 0:04:49.880
<v Speaker 4>All of our forecasts guidance that we showed zero, we

0:04:49.920 --> 0:04:52.599
<v Speaker 4>should start. That's the most important thing that she said.

0:04:52.880 --> 0:04:56.279
<v Speaker 4>She also said that effectively, China's a very important market

0:04:56.279 --> 0:04:59.240
<v Speaker 4>to us. It's very important to the United States, it's

0:04:59.320 --> 0:05:03.200
<v Speaker 4>very important to China. We would love the opportunity to

0:05:03.279 --> 0:05:07.080
<v Speaker 4>be able to re engage the Chinese market with excellent

0:05:07.120 --> 0:05:10.920
<v Speaker 4>products that we deliver and to be able to compete globally.

0:05:11.640 --> 0:05:15.520
<v Speaker 4>The Chinese market is very large. This year, my guess

0:05:15.600 --> 0:05:19.080
<v Speaker 4>is probably about fifty billion dollars. It's great for the

0:05:19.120 --> 0:05:22.799
<v Speaker 4>American people that we're able to compete in the Chinese market.

0:05:23.000 --> 0:05:25.359
<v Speaker 4>It's great for the China market that we're able to

0:05:25.360 --> 0:05:28.800
<v Speaker 4>provide and video's technology to them. It's great for the

0:05:28.839 --> 0:05:33.080
<v Speaker 4>rest of the world as Chinese software companies and Chinese

0:05:33.320 --> 0:05:36.719
<v Speaker 4>open source models leave China and are used all over

0:05:36.760 --> 0:05:39.880
<v Speaker 4>the world. And so I think it's fantastic that we're able.

0:05:40.200 --> 0:05:42.560
<v Speaker 4>It would be fantastic if we're able to participate in

0:05:42.560 --> 0:05:45.600
<v Speaker 4>the China market. But for now, we should just assume

0:05:46.080 --> 0:05:51.160
<v Speaker 4>the videos forecast for China market is zero. We're going

0:05:51.200 --> 0:05:54.240
<v Speaker 4>to continue to engage the US government, continue to engage

0:05:54.240 --> 0:05:58.200
<v Speaker 4>the China government to advise them and to encourage them

0:05:58.240 --> 0:05:59.680
<v Speaker 4>to allow us to go back and compete in the

0:05:59.680 --> 0:06:04.600
<v Speaker 4>open market. And so until then we should assume zero Jensen.

0:06:04.640 --> 0:06:08.240
<v Speaker 3>During the call, the US Commerce Department issued a statement

0:06:08.440 --> 0:06:11.960
<v Speaker 3>saying that you are now permitted to export up to

0:06:12.040 --> 0:06:16.280
<v Speaker 3>thirty five thousand Blackwell chips each to both saudiast Humane

0:06:16.640 --> 0:06:19.440
<v Speaker 3>and to the UAE through G forty two. But there

0:06:19.440 --> 0:06:22.680
<v Speaker 3>are some requirements that the US has a view, in

0:06:22.720 --> 0:06:28.600
<v Speaker 3>particular around controls of preventing tech transfer to China through

0:06:28.640 --> 0:06:30.479
<v Speaker 3>the Middle East. What can you tell us about your

0:06:30.560 --> 0:06:33.200
<v Speaker 3>understanding of what the US government's asking of you there

0:06:37.680 --> 0:06:38.359
<v Speaker 3>that that.

0:06:41.480 --> 0:06:45.359
<v Speaker 4>Element has been around for a long time is to

0:06:45.440 --> 0:06:50.200
<v Speaker 4>prevent diversion. Of course, over the years, people have speculated

0:06:50.240 --> 0:06:56.080
<v Speaker 4>about diversion. We've chased down every single concern, and we've

0:06:56.520 --> 0:07:01.800
<v Speaker 4>repeatedly tested and sampled data centers around the world and

0:07:02.160 --> 0:07:06.640
<v Speaker 4>found no diversion. And so this is an area that

0:07:07.360 --> 0:07:10.840
<v Speaker 4>will continue to be rigorous on. And there's a lot

0:07:10.840 --> 0:07:15.520
<v Speaker 4>of different ways to comply, and one of them, of course,

0:07:16.000 --> 0:07:19.400
<v Speaker 4>is to have it be run by American cloud. Another

0:07:19.480 --> 0:07:22.440
<v Speaker 4>way is just to make sure that we have measures

0:07:22.600 --> 0:07:26.040
<v Speaker 4>put in place, whether technology or processes, to ensure that

0:07:26.080 --> 0:07:27.040
<v Speaker 4>no diversion happens.

0:07:28.440 --> 0:07:30.800
<v Speaker 3>Jensen, The number one question I get for you is

0:07:30.880 --> 0:07:36.200
<v Speaker 3>always about energy. How severe is the energy shortage in

0:07:36.240 --> 0:07:39.000
<v Speaker 3>the context of AI expansion, and would you talk a

0:07:39.000 --> 0:07:43.080
<v Speaker 3>bit about power and whether power is a bigger constraint

0:07:43.600 --> 0:07:45.960
<v Speaker 3>for this buildout than the chips themselves.

0:07:49.760 --> 0:07:54.200
<v Speaker 4>When you're growing at the rate and scale of Nvidia,

0:07:55.120 --> 0:07:58.600
<v Speaker 4>Remember we're growing some sixty percent a year just quarter

0:07:58.720 --> 0:08:02.560
<v Speaker 4>to quarter growth of our company as ten billion dollars.

0:08:03.080 --> 0:08:05.520
<v Speaker 4>We grew an entire size of a company just from

0:08:05.680 --> 0:08:09.920
<v Speaker 4>in one quarter, and so the scale and the rate

0:08:10.000 --> 0:08:12.960
<v Speaker 4>at which we're growing everything's a challenge, which is the

0:08:12.960 --> 0:08:15.880
<v Speaker 4>reason why Nvidia has to be world class at our

0:08:15.920 --> 0:08:20.280
<v Speaker 4>supply chain, working with incredible providers and suppliers like TSMC

0:08:20.480 --> 0:08:23.720
<v Speaker 4>and the memory partners and all of our systems partners,

0:08:24.080 --> 0:08:29.520
<v Speaker 4>but also working downstream to work with energy providers, power

0:08:29.640 --> 0:08:34.600
<v Speaker 4>generator companies, all of the land power and shell providers,

0:08:34.920 --> 0:08:37.920
<v Speaker 4>so that we could make sure that as we launch

0:08:38.000 --> 0:08:41.280
<v Speaker 4>into the marketplace, as we deploy into the marketplace, land

0:08:41.320 --> 0:08:44.440
<v Speaker 4>power and shell will be ready for us. One of

0:08:44.480 --> 0:08:48.160
<v Speaker 4>our great advantages is that we have such a large

0:08:48.360 --> 0:08:52.160
<v Speaker 4>network of go to market. We're in every single cloud.

0:08:52.360 --> 0:08:54.679
<v Speaker 4>Every single cloud service provider is a customer of ours.

0:08:55.080 --> 0:08:59.000
<v Speaker 4>We're in every single GPU cloud, and so we have

0:08:59.040 --> 0:09:01.960
<v Speaker 4>a large network, not to mention OEMs, not to mention,

0:09:02.080 --> 0:09:06.000
<v Speaker 4>all around the world. Our customer base, our network of

0:09:06.040 --> 0:09:09.480
<v Speaker 4>partners is so large that we will find nooks and

0:09:09.520 --> 0:09:13.400
<v Speaker 4>crannies of power and large scale, medium scale, small scale

0:09:13.840 --> 0:09:16.400
<v Speaker 4>in different parts of the world. And so this is

0:09:16.440 --> 0:09:19.120
<v Speaker 4>a huge advantage of ours, and it stems from the

0:09:19.160 --> 0:09:25.280
<v Speaker 4>fact ed that Nvidia's architecture literally runs every model and today,

0:09:25.600 --> 0:09:28.920
<v Speaker 4>yesterday we announced a big news with Nentropic and so

0:09:29.040 --> 0:09:38.120
<v Speaker 4>now the premiere frontier models, open Ai, Anthropic, Xai, Gemini,

0:09:39.000 --> 0:09:44.600
<v Speaker 4>all the open sources, biological models, physical ai models, everything

0:09:44.679 --> 0:09:47.480
<v Speaker 4>in the world runs on Nvidia. And as a result

0:09:47.559 --> 0:09:51.480
<v Speaker 4>of that, irrespective of which cloud provider you are, it

0:09:51.600 --> 0:09:55.960
<v Speaker 4>is fantastic that we can deploy in your cloud because

0:09:55.960 --> 0:09:57.360
<v Speaker 4>the off take will be incredible.

0:09:58.280 --> 0:10:00.800
<v Speaker 3>Jensen, we can see where the hyperscale is are getting

0:10:00.840 --> 0:10:03.320
<v Speaker 3>the money, where they have the money to deploy and build.

0:10:03.800 --> 0:10:07.040
<v Speaker 3>But you mentioned Anthropic. With Anthropic or indeed open Ai,

0:10:07.360 --> 0:10:12.040
<v Speaker 3>they have tens of billions of dollars of commitments around

0:10:12.040 --> 0:10:14.560
<v Speaker 3>the world. Very simple like, how do you know the

0:10:14.600 --> 0:10:16.480
<v Speaker 3>open ai is good for it that it will be

0:10:16.520 --> 0:10:20.440
<v Speaker 3>able to find the money.

0:10:21.320 --> 0:10:27.559
<v Speaker 4>Well, we're thoughtful along with open Ai, thoughtful in aligning

0:10:27.640 --> 0:10:33.800
<v Speaker 4>on and taking into consideration visibility of demand and their

0:10:33.840 --> 0:10:37.599
<v Speaker 4>financing capabilities. All of that has to be in accordance,

0:10:38.080 --> 0:10:41.320
<v Speaker 4>has to be aligned, has to be coherent before we

0:10:41.360 --> 0:10:46.440
<v Speaker 4>start to build out. And so I think the ambitions large,

0:10:46.720 --> 0:10:50.680
<v Speaker 4>but the execution is disciplined, and that's really really important

0:10:50.720 --> 0:10:54.520
<v Speaker 4>to recognize we're very disciplined with our investment, We're disciplined

0:10:54.559 --> 0:10:58.840
<v Speaker 4>with our buildout. These are very large scale investments, and

0:10:58.920 --> 0:11:02.440
<v Speaker 4>so the two teams are quite disciplined, very disciplined in

0:11:02.480 --> 0:11:06.080
<v Speaker 4>thinking through the investment levels. Now, it's also important to

0:11:06.080 --> 0:11:09.679
<v Speaker 4>take a step back and realize that open AI anthropic.

0:11:10.280 --> 0:11:14.439
<v Speaker 4>These are the fastest growing company in the history of humanity.

0:11:14.920 --> 0:11:20.120
<v Speaker 4>Their off take, their end market demand is absolutely real

0:11:20.360 --> 0:11:23.960
<v Speaker 4>and absolutely incredible, and you could see that they're really

0:11:23.960 --> 0:11:26.640
<v Speaker 4>struggling to keep up with the demand that they have.

0:11:27.160 --> 0:11:29.840
<v Speaker 4>The engineering teams we work incredibly hard to make sure

0:11:29.840 --> 0:11:33.600
<v Speaker 4>that we bring them on more capacity, but also optimizing

0:11:33.640 --> 0:11:36.880
<v Speaker 4>their stack so that the usage of whatever capacity is

0:11:36.880 --> 0:11:40.800
<v Speaker 4>as efficient as possible. And meanwhile, there's so many new

0:11:40.920 --> 0:11:43.200
<v Speaker 4>use cases that they want to put back put out

0:11:43.200 --> 0:11:46.599
<v Speaker 4>into the world, and its currently limited by the capacity

0:11:46.600 --> 0:11:49.440
<v Speaker 4>they have, and so this is a really important time.

0:11:49.679 --> 0:11:53.000
<v Speaker 4>You're seeing an exponential growth in the amount of compute

0:11:53.000 --> 0:11:57.600
<v Speaker 4>demand necessary for AI, You're seeing an exponential growth of

0:11:57.920 --> 0:12:02.960
<v Speaker 4>adoption and use of AI, and the number of applications

0:12:02.960 --> 0:12:05.520
<v Speaker 4>that are going to be using these AI is also growing,

0:12:05.679 --> 0:12:07.880
<v Speaker 4>and so we've got to do our best to support

0:12:07.920 --> 0:12:11.040
<v Speaker 4>the scaling out of two of the most consequential companies

0:12:11.080 --> 0:12:14.360
<v Speaker 4>in history, and we're delied to be partnered with them.

0:12:15.360 --> 0:12:21.640
<v Speaker 3>Jensen investors have been worrying about depreciation. Software can actually

0:12:21.679 --> 0:12:24.320
<v Speaker 3>extend life like there are a one hundreds out there

0:12:24.559 --> 0:12:29.200
<v Speaker 3>in the real world still at full utilization. Are people

0:12:29.320 --> 0:12:33.360
<v Speaker 3>underestimating how long your chips stay useful or are they

0:12:33.440 --> 0:12:37.800
<v Speaker 3>kind of misunderstanding in the context of depreciation, how you're

0:12:37.840 --> 0:12:41.360
<v Speaker 3>handling generation to generation updates of GPU.

0:12:44.320 --> 0:12:51.560
<v Speaker 4>And Vidious architect and vidiot is unlike any other accelerator,

0:12:52.720 --> 0:12:57.640
<v Speaker 4>and the reason for that is because of Kuda's diversity

0:12:57.679 --> 0:13:01.800
<v Speaker 4>of capability and versatility. Remember I said two things earlier.

0:13:02.440 --> 0:13:06.880
<v Speaker 4>I said the fact that Nvidia participates and could accelerate

0:13:07.040 --> 0:13:11.000
<v Speaker 4>every phase of AI, pre training, post training, and inference.

0:13:11.440 --> 0:13:14.200
<v Speaker 4>We're the only architecture in the world that does that fantastically.

0:13:14.800 --> 0:13:18.160
<v Speaker 4>The second thing we do we run every single model,

0:13:19.120 --> 0:13:23.040
<v Speaker 4>and so most agentic systems, most clouds are running so

0:13:23.120 --> 0:13:26.760
<v Speaker 4>many different diverse type of models language models, vision models,

0:13:26.880 --> 0:13:30.319
<v Speaker 4>biological models, chemical models, for all the different fields of science.

0:13:31.040 --> 0:13:37.160
<v Speaker 4>Nvidia could be used across the entire lifespan of the technology.

0:13:37.559 --> 0:13:40.200
<v Speaker 4>And so if you look at products that we shipped

0:13:40.400 --> 0:13:44.200
<v Speaker 4>amper A one hundred we shipped six years ago. But

0:13:44.280 --> 0:13:47.160
<v Speaker 4>because we're continuing, our installed bas is so large, our

0:13:47.200 --> 0:13:51.920
<v Speaker 4>diversity is so great, we could continuously update our software

0:13:52.440 --> 0:13:54.960
<v Speaker 4>bring value to our customers on the one hand, but

0:13:55.040 --> 0:13:59.640
<v Speaker 4>because our versatility is so great based on the capability

0:13:59.679 --> 0:14:02.160
<v Speaker 4>they need, they could use our GPUs for a very

0:14:02.240 --> 0:14:05.000
<v Speaker 4>very long time. Now, remember A one hundred and six

0:14:05.160 --> 0:14:08.640
<v Speaker 4>years old. However, it is still an order of a

0:14:08.679 --> 0:14:12.240
<v Speaker 4>magnitude faster than any CPU could put the bear, so

0:14:12.360 --> 0:14:14.760
<v Speaker 4>it is still the best computer. It is still the

0:14:14.800 --> 0:14:17.920
<v Speaker 4>best processor for much of the workload in the cloud,

0:14:18.480 --> 0:14:24.440
<v Speaker 4>and most people misunderstand that because unlike US, most accelerators

0:14:25.000 --> 0:14:28.720
<v Speaker 4>are kind of singular use, because they don't have diversity,

0:14:29.000 --> 0:14:31.800
<v Speaker 4>because they don't have versatility, because they're not great at

0:14:31.840 --> 0:14:35.720
<v Speaker 4>every phase of AI. Once they're used for whatever they

0:14:35.720 --> 0:14:39.000
<v Speaker 4>were designed to do, their value falls off a cliff.

0:14:39.280 --> 0:14:42.120
<v Speaker 4>That is not true with Nvidio Jensen.

0:14:42.160 --> 0:14:45.520
<v Speaker 3>My final question is a point of clarification, if I may.

0:14:45.600 --> 0:14:48.800
<v Speaker 3>You were asked on the call about content and in

0:14:48.920 --> 0:14:52.520
<v Speaker 3>video's contribution to any given piece of AI infrastructure, and

0:14:52.600 --> 0:14:55.320
<v Speaker 3>what you said was Hopper around twenty to twenty five

0:14:55.800 --> 0:15:00.360
<v Speaker 3>black COO about thirty With those figures billions of dollars

0:15:00.640 --> 0:15:04.480
<v Speaker 3>in dollar terms on a one gigawa data center, are

0:15:04.480 --> 0:15:06.800
<v Speaker 3>you talking about percentages of total cost?

0:15:10.280 --> 0:15:10.520
<v Speaker 1>Oh?

0:15:10.680 --> 0:15:15.800
<v Speaker 4>Thank you, billions of dollars, billions of dollars. Yeah, And

0:15:15.880 --> 0:15:20.600
<v Speaker 4>so for our vera Rubens system, a one gigawak data

0:15:20.640 --> 0:15:23.480
<v Speaker 4>center is probably something along the lines of fifty to

0:15:23.520 --> 0:15:27.320
<v Speaker 4>fifty five, and Video's contribution is probably about thirty five

0:15:27.360 --> 0:15:28.320
<v Speaker 4>of that.

0:15:28.320 --> 0:15:31.600
<v Speaker 2>That was in Nvidia CEO Jensen Wong speaking to Bloomberg's

0:15:31.720 --> 0:15:34.280
<v Speaker 2>Ed Ludlow, bringing it to you here on the Daybreak

0:15:34.280 --> 0:15:44.440
<v Speaker 2>Asia podcast. Welcome back to the Daybreak Asia podcast. I'm

0:15:44.480 --> 0:15:48.560
<v Speaker 2>Doug Prisoner. On Wednesday, President Trump renewed his attacks against

0:15:48.560 --> 0:15:51.240
<v Speaker 2>FED Shair J. Powell. Trump said he would love to

0:15:51.360 --> 0:15:54.200
<v Speaker 2>fire him for not lowering interest rates as much as

0:15:54.200 --> 0:15:57.560
<v Speaker 2>the President would like, and Trump suggested he would fire

0:15:57.600 --> 0:16:01.200
<v Speaker 2>Scott Besant if the Treasury Secretary doesn't help secure lower

0:16:01.280 --> 0:16:05.400
<v Speaker 2>interest rates well. Importantly, the Treasury Secretary has no influence

0:16:05.480 --> 0:16:09.200
<v Speaker 2>on the FED, although Best is leading the selection process

0:16:09.240 --> 0:16:12.120
<v Speaker 2>for the next FED share to replace Powell when his

0:16:12.240 --> 0:16:15.400
<v Speaker 2>term expires in May twenty twenty six. Also today we

0:16:15.440 --> 0:16:17.680
<v Speaker 2>got minutes from the last FED meeting, and they were

0:16:17.720 --> 0:16:22.400
<v Speaker 2>a little hawkish. Policymakers were divided. Several were against lowering

0:16:22.480 --> 0:16:24.960
<v Speaker 2>the Fed's benchmark rate last month, and many said it

0:16:25.000 --> 0:16:27.840
<v Speaker 2>would likely be appropriate to keep rate steady for the

0:16:27.840 --> 0:16:31.320
<v Speaker 2>remainder of this year. So one of the hallmarks of

0:16:31.400 --> 0:16:33.960
<v Speaker 2>the US session was this idea that we have to

0:16:34.000 --> 0:16:37.520
<v Speaker 2>reduce bets on a December rate cut. For a closer look, now,

0:16:37.560 --> 0:16:39.960
<v Speaker 2>I'm joined by Clayton Trick. He is head of portfolio

0:16:40.080 --> 0:16:43.880
<v Speaker 2>management at angel Oak Capital Advisors. Clayton is on the

0:16:43.880 --> 0:16:46.560
<v Speaker 2>line from Atlanta, Georgia. Thank you, sir for making time

0:16:46.560 --> 0:16:48.800
<v Speaker 2>to chat with me. So I'm looking at the swaps

0:16:48.840 --> 0:16:51.960
<v Speaker 2>market pricing right now. Thirty percent probability that we're going

0:16:52.000 --> 0:16:55.160
<v Speaker 2>to get a twenty five basis point rate cut in December.

0:16:55.240 --> 0:16:56.880
<v Speaker 2>Is that how you see things right now?

0:16:57.640 --> 0:17:02.440
<v Speaker 5>Yes, we do. We feel that the press conference right

0:17:02.480 --> 0:17:05.040
<v Speaker 5>after the last meeting, you saw Chim and Pale really

0:17:05.080 --> 0:17:07.960
<v Speaker 5>push against the market. You know, before that you had

0:17:08.000 --> 0:17:10.160
<v Speaker 5>one hundred percent, you know, mere one hundred percent chance

0:17:10.200 --> 0:17:12.960
<v Speaker 5>of a rate cut for the December meeting, and he

0:17:13.040 --> 0:17:15.280
<v Speaker 5>really wanted to push against the market and really keep

0:17:15.280 --> 0:17:20.400
<v Speaker 5>the optionality open. You know. After that we saw continuous

0:17:20.400 --> 0:17:24.439
<v Speaker 5>public appearances from more hawkish and centrist type members, you know,

0:17:24.480 --> 0:17:27.560
<v Speaker 5>thinking you of a FED presidence like Schmid, Logan, Hammock,

0:17:27.600 --> 0:17:30.119
<v Speaker 5>as well as centrists like Bostick and Collins that really

0:17:30.359 --> 0:17:33.679
<v Speaker 5>kind of reiterated their apprehension toward a December cut. So

0:17:34.200 --> 0:17:37.080
<v Speaker 5>not surprising the market is pricing in a much lower

0:17:37.119 --> 0:17:39.879
<v Speaker 5>probability of a cut. But that being said, we haven't

0:17:39.920 --> 0:17:42.560
<v Speaker 5>had a lot of economic data obviously the last couple

0:17:42.960 --> 0:17:45.640
<v Speaker 5>weeks with the government shutdown, so things could definitely change

0:17:45.680 --> 0:17:46.600
<v Speaker 5>here in the coming weeks.

0:17:46.840 --> 0:17:49.119
<v Speaker 2>We also heard today from Kevin Hassett to the Director

0:17:49.160 --> 0:17:51.960
<v Speaker 2>of the National Economic Council, and Hassett, as we know,

0:17:52.000 --> 0:17:54.000
<v Speaker 2>as one of the finals to become the next FED

0:17:54.040 --> 0:17:57.800
<v Speaker 2>share I found it interesting that today he was discussing

0:17:57.880 --> 0:18:00.840
<v Speaker 2>consumer price pressures and saying that they may not be

0:18:00.920 --> 0:18:04.720
<v Speaker 2>fully tamed up until this point when I've heard Hasset speak,

0:18:05.119 --> 0:18:08.080
<v Speaker 2>there's been a dubbish quality today that seemed to tilt

0:18:08.160 --> 0:18:11.239
<v Speaker 2>a little on the hawkish side. So Clayton, can you

0:18:11.280 --> 0:18:14.120
<v Speaker 2>give me your sense of how you view the inflation

0:18:14.200 --> 0:18:15.000
<v Speaker 2>story right now?

0:18:16.359 --> 0:18:19.200
<v Speaker 5>Yeah, we have continued to believe that, you know, core

0:18:19.320 --> 0:18:21.720
<v Speaker 5>goods inflation was going to put a little bit of

0:18:21.760 --> 0:18:24.640
<v Speaker 5>upward pressure with tariffs, and it was going to take

0:18:24.720 --> 0:18:29.600
<v Speaker 5>some time. That being said, thirty five percent of core PCE,

0:18:30.640 --> 0:18:33.399
<v Speaker 5>thirty five percent of core CPI is shelter, and that

0:18:33.480 --> 0:18:36.000
<v Speaker 5>has continued to decelerate. That's what we've got in the

0:18:36.040 --> 0:18:39.840
<v Speaker 5>most recent number, the figure that we got within CPI

0:18:39.960 --> 0:18:43.520
<v Speaker 5>during the shutdown, and we expect home prices to continue

0:18:43.560 --> 0:18:46.840
<v Speaker 5>to decelebrate or fall, which will continue to bring down

0:18:47.200 --> 0:18:51.360
<v Speaker 5>the shelter component of inflation. On the good side, that

0:18:51.359 --> 0:18:53.480
<v Speaker 5>that's going to take a lot longer to flow through.

0:18:53.920 --> 0:18:56.119
<v Speaker 5>That being said, we're already getting some rhetoric from from

0:18:56.240 --> 0:18:58.879
<v Speaker 5>DC talking about how they may be rolling back some

0:18:58.960 --> 0:19:03.000
<v Speaker 5>of the tariffs on specific things that are having to

0:19:03.040 --> 0:19:06.560
<v Speaker 5>do with consumer spending. So we could see kind of

0:19:06.560 --> 0:19:11.359
<v Speaker 5>a slight rollback of tariffs into next year. Shelter is

0:19:11.359 --> 0:19:14.800
<v Speaker 5>going to continue to put downward pressure on core inflation,

0:19:14.960 --> 0:19:17.199
<v Speaker 5>and so we don't think inflation is really going to

0:19:17.280 --> 0:19:20.439
<v Speaker 5>rise much from here, but could see remaining in the

0:19:20.520 --> 0:19:24.840
<v Speaker 5>high twos. So FED funds now in the high threes.

0:19:24.960 --> 0:19:26.640
<v Speaker 5>We think there is room for them to be looking

0:19:26.680 --> 0:19:29.320
<v Speaker 5>to cut but I'm not surprised you're getting a little

0:19:29.359 --> 0:19:31.440
<v Speaker 5>bit of pushback in the most recent comments.

0:19:31.480 --> 0:19:34.439
<v Speaker 2>So, if you're expecting some more easing and you're not

0:19:34.560 --> 0:19:37.600
<v Speaker 2>expecting inflation to be a problem, what's your investment strategy

0:19:37.680 --> 0:19:38.080
<v Speaker 2>right now?

0:19:39.320 --> 0:19:42.680
<v Speaker 5>Yeah, so there are significant opportunities, especially with recent and

0:19:42.680 --> 0:19:45.240
<v Speaker 5>the recent pullback, we've been keeping some dry powder on

0:19:45.280 --> 0:19:48.800
<v Speaker 5>the sidelines, really looking to step into areas like agency

0:19:49.040 --> 0:19:52.000
<v Speaker 5>and non agency mortgage backed securities. There's been a lot

0:19:52.000 --> 0:19:55.600
<v Speaker 5>of comments recently regarding how corporate credit is on the

0:19:55.640 --> 0:19:58.040
<v Speaker 5>tighter end of the range. You know, really that market's

0:19:58.080 --> 0:20:00.919
<v Speaker 5>highly correlated to where equity valuations are. But outside of

0:20:00.960 --> 0:20:04.320
<v Speaker 5>corporate credit, thinking of the securitized bond market, that areas

0:20:04.440 --> 0:20:08.359
<v Speaker 5>is still actually really wide compared to corporate spreads thinking

0:20:08.440 --> 0:20:10.320
<v Speaker 5>back to twenty twenty one and spreads, we're a lot

0:20:10.400 --> 0:20:15.160
<v Speaker 5>tighter across the market. You're significantly wider in securitized bonds

0:20:15.240 --> 0:20:18.119
<v Speaker 5>versus investment grade and higher corporate. So we see that

0:20:18.160 --> 0:20:20.520
<v Speaker 5>as an attracted area of the market. And from a

0:20:20.560 --> 0:20:23.600
<v Speaker 5>credit perspective, the housing market we view is actually still

0:20:23.640 --> 0:20:26.480
<v Speaker 5>in very good shape, and so mortgage bond should huddle

0:20:26.600 --> 0:20:30.560
<v Speaker 5>pulled up well in either continued muddling along the economy

0:20:31.000 --> 0:20:33.359
<v Speaker 5>or you know, an environment where the FED needs to

0:20:33.400 --> 0:20:36.160
<v Speaker 5>actually maybe be cutting more because the layer market is

0:20:36.440 --> 0:20:37.359
<v Speaker 5>turning in the other direction.

0:20:37.480 --> 0:20:39.679
<v Speaker 2>What about on the equity side, Clayton, I'm curious to

0:20:39.920 --> 0:20:42.399
<v Speaker 2>know where you're focused in the equity space.

0:20:43.920 --> 0:20:46.520
<v Speaker 5>Yeah, so we're continuing to watch obviously the moves impact

0:20:46.520 --> 0:20:50.360
<v Speaker 5>that's very important for the coming growth and data centers

0:20:50.440 --> 0:20:54.080
<v Speaker 5>and anything around AI led cap X. That being said,

0:20:54.280 --> 0:20:56.479
<v Speaker 5>we think it's actually very notable to continue to be

0:20:56.600 --> 0:21:01.800
<v Speaker 5>watching performance of public BDCs. Tinded to see discussions and

0:21:01.840 --> 0:21:04.919
<v Speaker 5>concerns around middle market lending about the growth and private

0:21:04.960 --> 0:21:09.639
<v Speaker 5>credit public market publicly traded BDCs is an area that

0:21:09.640 --> 0:21:13.240
<v Speaker 5>we can continue to get insight into that area of

0:21:13.280 --> 0:21:17.240
<v Speaker 5>the market. We've seen that really decouple from SMP and

0:21:17.320 --> 0:21:20.360
<v Speaker 5>tech performance really since the second quarter, and so we're

0:21:20.359 --> 0:21:23.119
<v Speaker 5>really keeping a keen focus heading into the year to

0:21:23.160 --> 0:21:27.280
<v Speaker 5>see where those the valuations within that market looks heading

0:21:27.320 --> 0:21:28.160
<v Speaker 5>into twenty twenty six.

0:21:28.359 --> 0:21:33.520
<v Speaker 2>And what about opportunities offshore? Are you finding any.

0:21:33.359 --> 0:21:36.480
<v Speaker 5>We definitely think of vests should be thinking globally right now.

0:21:38.160 --> 0:21:41.200
<v Speaker 5>US acceptionalism in the last few years has really stuck out,

0:21:41.200 --> 0:21:44.159
<v Speaker 5>and now you're seeing global opportunities coming back from an

0:21:44.200 --> 0:21:46.480
<v Speaker 5>equity perspective. We think that's still going to be the case,

0:21:46.520 --> 0:21:49.800
<v Speaker 5>and so you know, we're we're recommending investors really focus

0:21:50.040 --> 0:21:53.919
<v Speaker 5>on global diversification, not just US equities heading into next year.

0:21:53.960 --> 0:21:55.720
<v Speaker 2>All right, Clayton, we'll leave it there. Thank you so much.

0:21:55.760 --> 0:21:59.679
<v Speaker 2>Clayton Trick, head of portfolio management at Angel Oak Capital Advisors,

0:22:00.119 --> 0:22:02.400
<v Speaker 2>on the line from Atlanta, Georgia, here on the Daybreak

0:22:02.440 --> 0:22:07.520
<v Speaker 2>Asia Podcast. Thanks for listening to today's episode of the

0:22:07.560 --> 0:22:11.760
<v Speaker 2>Bloomberg Daybreak Asia Edition podcast. Each weekday, we look at

0:22:11.760 --> 0:22:16.240
<v Speaker 2>the story shaping markets, finance, and geopolitics in the Asia Pacific.

0:22:16.480 --> 0:22:19.760
<v Speaker 2>You can find us on Apple, Spotify, the Bloomberg Podcast

0:22:19.840 --> 0:22:23.200
<v Speaker 2>YouTube channel, or anywhere else you listen. Join us again

0:22:23.240 --> 0:22:26.520
<v Speaker 2>tomorrow for insight on the market moves from Hong Kong

0:22:26.640 --> 0:22:31.040
<v Speaker 2>to Singapore and Australia. I'm Doug Prisner, and this is

0:22:31.080 --> 0:22:31.639
<v Speaker 2>Bloomberg