WEBVTT - Arm CEO Rene Haas Talks Earnings and AI Outlook

0:00:00.040 --> 0:00:01.800
<v Speaker 1>We're of course going to be talking about the chip

0:00:01.800 --> 0:00:05.200
<v Speaker 1>design firm ARM, which has really bounced off of its

0:00:05.200 --> 0:00:07.320
<v Speaker 1>lows in terms of its share price throughout the trading

0:00:07.360 --> 0:00:09.600
<v Speaker 1>of today. We're currently over the last two days down

0:00:09.600 --> 0:00:11.640
<v Speaker 1>by one point three percent. Coming out after the bell

0:00:11.680 --> 0:00:14.680
<v Speaker 1>with its earnings tepid forecast ed was what the market

0:00:14.760 --> 0:00:17.279
<v Speaker 1>seemed to be focusing in on, even though they absolutely

0:00:17.320 --> 0:00:19.800
<v Speaker 1>smashed it in terms of their fiscal fourth quarter numbers

0:00:19.840 --> 0:00:22.720
<v Speaker 1>and their fourth first quarter look ahead. Let's stick in

0:00:22.760 --> 0:00:25.400
<v Speaker 1>to some of that caution with Renee. Has is the

0:00:25.520 --> 0:00:28.720
<v Speaker 1>ARM CEO, Renee wonderful to have time with you. And look,

0:00:28.800 --> 0:00:31.080
<v Speaker 1>there does seem to be a worry about your full

0:00:31.160 --> 0:00:32.280
<v Speaker 1>year forecasts?

0:00:32.600 --> 0:00:33.640
<v Speaker 2>Are you being cautious?

0:00:35.560 --> 0:00:37.560
<v Speaker 3>Well, thanks both for having me Ed and Carroll. And

0:00:38.080 --> 0:00:41.199
<v Speaker 3>we just came off a record year in terms of revenue.

0:00:41.200 --> 0:00:43.320
<v Speaker 3>We were up twenty percent a little bit over twenty

0:00:43.360 --> 0:00:48.199
<v Speaker 3>percent our fiscal twenty three from twenty twenty two, and

0:00:48.240 --> 0:00:51.400
<v Speaker 3>we're actually forecasting even higher growth this year, north of

0:00:51.440 --> 0:00:52.159
<v Speaker 3>twenty percent.

0:00:52.320 --> 0:00:54.040
<v Speaker 4>And we also signaled to.

0:00:53.960 --> 0:00:57.960
<v Speaker 3>The markets yesterday that in twenty five, twenty six, twenty seven,

0:00:58.520 --> 0:01:02.320
<v Speaker 3>we see that growth continuing. So we have incredible visibility

0:01:02.600 --> 0:01:05.120
<v Speaker 3>into our business and we're very very confident of this

0:01:05.160 --> 0:01:06.240
<v Speaker 3>growth rate going forward.

0:01:09.520 --> 0:01:12.320
<v Speaker 2>We're just seeing your shares actually ticking to positive territory

0:01:12.319 --> 0:01:16.319
<v Speaker 2>rene up now six tenths of one percent. The underlying

0:01:16.400 --> 0:01:19.800
<v Speaker 2>story is the build out in AI infrastructure. Right, we're

0:01:19.800 --> 0:01:25.760
<v Speaker 2>talking about data center powered GP by GPUs. Your numbers

0:01:25.760 --> 0:01:29.400
<v Speaker 2>were good. Tell me about the underlying demand, then about

0:01:29.400 --> 0:01:32.000
<v Speaker 2>the long term and the addressable market you think is

0:01:32.040 --> 0:01:33.399
<v Speaker 2>either intact or is not.

0:01:35.120 --> 0:01:38.840
<v Speaker 3>Well. I think this AI buildout, as you describe, or

0:01:38.880 --> 0:01:42.480
<v Speaker 3>maybe said another way, just expanding capacity to run these

0:01:42.480 --> 0:01:44.720
<v Speaker 3>foundation models, to do more and more training, to do

0:01:44.760 --> 0:01:47.760
<v Speaker 3>more and more inference. We really are only at the

0:01:47.840 --> 0:01:50.880
<v Speaker 3>very beginning because when you start to think about the

0:01:50.960 --> 0:01:57.400
<v Speaker 3>capabilities that this could unleash, whether it's around healthcare, pharmer research, productivity, gains,

0:01:57.480 --> 0:02:01.240
<v Speaker 3>call centers, we're still in the very very early days.

0:02:01.600 --> 0:02:03.920
<v Speaker 3>That all starts with having to do this level of

0:02:03.960 --> 0:02:06.640
<v Speaker 3>training and inference in the cloud, but it ultimately will

0:02:06.640 --> 0:02:10.239
<v Speaker 3>find itself in every single edge device, whether that's a PC,

0:02:11.400 --> 0:02:15.240
<v Speaker 3>your smartphone, your car, and whether it's all those devices

0:02:15.240 --> 0:02:18.079
<v Speaker 3>I've mentioned from the data center to the edge devices,

0:02:18.080 --> 0:02:21.400
<v Speaker 3>they all run on ARM. So we have incredible visibility

0:02:21.440 --> 0:02:23.480
<v Speaker 3>to where this is all going, which is why we're

0:02:23.560 --> 0:02:24.240
<v Speaker 3>very confident.

0:02:24.000 --> 0:02:27.080
<v Speaker 4>In the growth rates. They're also one of the big problems.

0:02:26.760 --> 0:02:29.679
<v Speaker 3>You've got with all of these AI data centers is

0:02:29.720 --> 0:02:33.120
<v Speaker 3>around energy and power, so power efficiency being so key,

0:02:33.200 --> 0:02:36.320
<v Speaker 3>it's what ARM is really good at. Increasingly we're seeing

0:02:36.320 --> 0:02:39.919
<v Speaker 3>the most complex applications moving to ARM and most sophisticated

0:02:39.960 --> 0:02:44.120
<v Speaker 3>training ship on the planet that was just announced Grace Blackwell,

0:02:44.160 --> 0:02:49.240
<v Speaker 3>well that's based on ARM.

0:02:49.400 --> 0:02:51.440
<v Speaker 1>Okay, So you're managing to really think that you're going

0:02:51.480 --> 0:02:55.200
<v Speaker 1>to be the server play as well as the PC play,

0:02:55.200 --> 0:02:56.840
<v Speaker 1>the cell phone play, and I want to focus in

0:02:56.880 --> 0:02:58.839
<v Speaker 1>on the cell phone play, Renee, because that's been where

0:02:58.840 --> 0:03:01.640
<v Speaker 1>your bread and muster has been history. How are we

0:03:01.680 --> 0:03:04.800
<v Speaker 1>looking from a smartphone perspective? Is the market looking strong

0:03:04.880 --> 0:03:07.200
<v Speaker 1>to you? We've had many a mixed message coming from

0:03:07.240 --> 0:03:08.400
<v Speaker 1>China to mod for example.

0:03:10.280 --> 0:03:13.080
<v Speaker 3>Overall, what we've seen the smartphone market briefly for ARM

0:03:13.160 --> 0:03:15.960
<v Speaker 3>has been quite a good growth rate in terms of royalties.

0:03:16.160 --> 0:03:19.399
<v Speaker 3>Our version nine which is now being used in many

0:03:19.440 --> 0:03:22.680
<v Speaker 3>of the premium mobile phones, that drives a higher royalty

0:03:22.760 --> 0:03:26.080
<v Speaker 3>rate for ARM. There's also more complex CPUs that go

0:03:26.200 --> 0:03:28.040
<v Speaker 3>into that that's also better.

0:03:27.800 --> 0:03:29.959
<v Speaker 4>For ARM and going forward carrolling.

0:03:30.000 --> 0:03:31.760
<v Speaker 3>One of the things that we're seeing, and it's not

0:03:31.840 --> 0:03:35.600
<v Speaker 3>just in smartphones, is that as these AI models are

0:03:35.600 --> 0:03:39.680
<v Speaker 3>moving so fast, the hardware can't keep up with the software.

0:03:40.080 --> 0:03:44.000
<v Speaker 3>The software innovation is happening so quickly that by the

0:03:44.040 --> 0:03:46.760
<v Speaker 3>time the hardware is ready to run those models, everyone

0:03:46.800 --> 0:03:50.320
<v Speaker 3>wishes they had more performance, they had more efficiency. So

0:03:50.360 --> 0:03:52.840
<v Speaker 3>what does that mean for ARM. It's driving growth in

0:03:52.880 --> 0:03:55.960
<v Speaker 3>our licensing activity. People are looking to do more and

0:03:56.000 --> 0:04:00.000
<v Speaker 3>more design ships faster and faster, and that's all good

0:04:00.080 --> 0:04:02.320
<v Speaker 3>for us going forward. So I think going forward you're

0:04:02.320 --> 0:04:04.200
<v Speaker 3>going to see more and more innovation happening, not only

0:04:04.200 --> 0:04:07.280
<v Speaker 3>in the smartphones, but across all these edge devices.

0:04:10.400 --> 0:04:12.600
<v Speaker 1>What's interesting in AY is it's hard to keep up

0:04:12.640 --> 0:04:16.080
<v Speaker 1>with the pace of geopolitical change as well. The latest

0:04:16.080 --> 0:04:18.560
<v Speaker 1>news coming that Huawei of course is not going to

0:04:18.560 --> 0:04:21.560
<v Speaker 1>have access to qualcom to Intel chips. You were, of

0:04:21.560 --> 0:04:25.880
<v Speaker 1>course a UK based company, but are affected by US policies.

0:04:26.440 --> 0:04:29.720
<v Speaker 1>Has this impacted your business the limitations of Huawei's access

0:04:29.760 --> 0:04:33.200
<v Speaker 1>to chip designed to chip technology to licenses.

0:04:34.880 --> 0:04:37.479
<v Speaker 3>Yeah, So that issue that you referred to specifically was

0:04:37.880 --> 0:04:40.280
<v Speaker 3>when Huawei was placed on the entity list I think

0:04:40.320 --> 0:04:43.839
<v Speaker 3>twenty nineteen, twenty twenty, companies had to apply for a

0:04:43.880 --> 0:04:47.719
<v Speaker 3>license to exempt them to ship to Huawei. So a

0:04:47.760 --> 0:04:51.919
<v Speaker 3>number of companies asked for those licenses, they got those licenses.

0:04:52.040 --> 0:04:54.840
<v Speaker 3>Now those licenses are being revoked. We don't follow in

0:04:54.920 --> 0:04:57.400
<v Speaker 3>that category in any way, shape or form. We didn't

0:04:57.400 --> 0:05:00.520
<v Speaker 3>apply for any licenses at the time to ship, complied

0:05:00.520 --> 0:05:02.880
<v Speaker 3>with the export control as they were laid out. So

0:05:03.160 --> 0:05:06.120
<v Speaker 3>there's really anne event for us in terms of what

0:05:06.160 --> 0:05:07.800
<v Speaker 3>you're seeing with Qualcom and or Intel.

0:05:11.240 --> 0:05:14.200
<v Speaker 2>We are speaking live to the Arms CEO, Renee has

0:05:14.279 --> 0:05:17.600
<v Speaker 2>We're on the ground here at Bloomberg Tech in San Francisco.

0:05:17.960 --> 0:05:20.520
<v Speaker 2>Last week, Renee Christiano and Mom was on the show

0:05:20.600 --> 0:05:24.240
<v Speaker 2>telling Caroline and I, this is the year of the AIPC.

0:05:25.080 --> 0:05:27.360
<v Speaker 2>You were asked about that on your earning school last

0:05:27.440 --> 0:05:30.640
<v Speaker 2>night and you gave a slightly different answer. And maybe

0:05:30.640 --> 0:05:33.799
<v Speaker 2>it's not the year of the AIPC more the twelve

0:05:33.880 --> 0:05:36.560
<v Speaker 2>to thirty six month window. And you don't want to

0:05:36.600 --> 0:05:39.160
<v Speaker 2>see just one PC supplier, you said you'd like to

0:05:39.200 --> 0:05:43.760
<v Speaker 2>see two or three. What's your beef with Qualcom?

0:05:43.880 --> 0:05:44.000
<v Speaker 4>Now?

0:05:44.000 --> 0:05:48.120
<v Speaker 3>When I look at the PC ecosystem, one large ecosystem

0:05:48.160 --> 0:05:50.080
<v Speaker 3>has already moved to ARM in a very big way.

0:05:51.080 --> 0:05:54.080
<v Speaker 3>Apple is now one percent based on ARM. All the

0:05:54.120 --> 0:05:57.560
<v Speaker 3>Apple silicon is based on ARM and you see amazingly

0:05:57.600 --> 0:06:02.520
<v Speaker 3>good products relative to what they've delivered, fantastic battery life, performance,

0:06:02.600 --> 0:06:05.800
<v Speaker 3>thin and light, no fans. When you think about the

0:06:05.800 --> 0:06:08.840
<v Speaker 3>Windows market, it's a very different market. It's highly fragmented.

0:06:08.880 --> 0:06:11.880
<v Speaker 3>You have lots of different players. The ecosystem matters, the

0:06:12.000 --> 0:06:16.960
<v Speaker 3>channel matters, price points matter, high end gaming machines versus

0:06:17.920 --> 0:06:21.560
<v Speaker 3>low end devices that are like cloud enabled. So what

0:06:21.560 --> 0:06:25.400
<v Speaker 3>does all of that mean. It generally has meant that breadth,

0:06:25.920 --> 0:06:29.719
<v Speaker 3>vendor choice, multiple options to provide a full scope is

0:06:29.720 --> 0:06:33.080
<v Speaker 3>what matters. And what I'm hearing is over the next

0:06:33.080 --> 0:06:35.120
<v Speaker 3>couple of years, the Windows ecosystem is going to be

0:06:35.160 --> 0:06:35.920
<v Speaker 3>able to afford that.

0:06:36.480 --> 0:06:37.839
<v Speaker 4>And I think over.

0:06:37.720 --> 0:06:41.279
<v Speaker 3>The next two three years, I do believe Windows Unarmed

0:06:41.760 --> 0:06:44.480
<v Speaker 3>will be real. I think you'll see multiple players, multiple

0:06:44.520 --> 0:06:47.600
<v Speaker 3>price points, multiple units, and I think you'll see meaningful

0:06:47.640 --> 0:06:50.320
<v Speaker 3>market share that we start to gain the kind of

0:06:50.360 --> 0:06:52.360
<v Speaker 3>performance you see in the other ecosystem. I think we'll

0:06:52.360 --> 0:07:01.640
<v Speaker 3>find its way into the Windows ecosystem.

0:06:57.800 --> 0:06:59.840
<v Speaker 2>All Rennie, I wanted to talk about geography really quick

0:07:00.120 --> 0:07:02.440
<v Speaker 2>we're here in San Francisco, right, there's a lot about

0:07:02.760 --> 0:07:07.000
<v Speaker 2>Americas are in d focus on AI related chips. Are

0:07:07.040 --> 0:07:11.280
<v Speaker 2>you seeing this sort of equivalent activity in Europe, for example,

0:07:11.400 --> 0:07:13.600
<v Speaker 2>any of your customers outside of those markets.

0:07:14.680 --> 0:07:16.480
<v Speaker 3>Yeah, well I'm in San Francisco today too, so I

0:07:16.480 --> 0:07:17.600
<v Speaker 3>will see you a little bit later.

0:07:17.720 --> 0:07:20.680
<v Speaker 4>But in general, I think the.

0:07:20.680 --> 0:07:24.960
<v Speaker 3>Geopolitics are something that all tech CEOs are now having

0:07:25.000 --> 0:07:29.720
<v Speaker 3>to figure out and work with AI models, foundation models,

0:07:30.360 --> 0:07:33.920
<v Speaker 3>sovereign clouds, thinking about what level of training takes place

0:07:34.680 --> 0:07:38.440
<v Speaker 3>in a country versus outside the country, where the weights sit,

0:07:38.520 --> 0:07:40.480
<v Speaker 3>et cetera. That these are all the kind of things

0:07:40.480 --> 0:07:43.320
<v Speaker 3>that politicians have never really had to think about in

0:07:43.360 --> 0:07:43.800
<v Speaker 3>the past.

0:07:43.880 --> 0:07:45.440
<v Speaker 4>So we're involved in a lot.

0:07:45.280 --> 0:07:49.160
<v Speaker 3>Of those conversations, whether that's in the United States, whether

0:07:49.160 --> 0:07:51.640
<v Speaker 3>that's in Europe, and really just trying to understand it

0:07:51.640 --> 0:07:55.280
<v Speaker 3>because any lawmakers in all these jurisdictions are just trying

0:07:55.320 --> 0:07:57.400
<v Speaker 3>to figure it all out. And as I mentioned before,

0:07:57.960 --> 0:08:00.400
<v Speaker 3>as the software and models are moving so fast asked,

0:08:00.960 --> 0:08:03.320
<v Speaker 3>it's difficult for everyone to keep up, But we are

0:08:03.480 --> 0:08:05.000
<v Speaker 3>central to all those discussions.

0:08:08.640 --> 0:08:11.240
<v Speaker 1>Renee, what's been keeping up is your valuation?

0:08:11.760 --> 0:08:12.000
<v Speaker 4>Boy?

0:08:12.600 --> 0:08:15.400
<v Speaker 1>I mean, do you think there's too much exuberance around

0:08:15.440 --> 0:08:18.880
<v Speaker 1>AI valuations out there. Are you going to make the

0:08:18.960 --> 0:08:20.880
<v Speaker 1>most of it by well, we talked to one point

0:08:20.880 --> 0:08:22.160
<v Speaker 1>of listing in the UK too.

0:08:24.080 --> 0:08:26.400
<v Speaker 3>Yeah, you know, I don't think about the valuations as

0:08:26.480 --> 0:08:28.920
<v Speaker 3>much as I just think about the AI opportunity, which

0:08:29.000 --> 0:08:33.720
<v Speaker 3>I frankly believe is undercalled in terms of just what

0:08:33.760 --> 0:08:36.280
<v Speaker 3>it's going to mean relative to society and what it

0:08:36.320 --> 0:08:39.320
<v Speaker 3>can do for our planet. I think again we are

0:08:39.360 --> 0:08:42.760
<v Speaker 3>in very very early days in terms of the capabilities

0:08:42.800 --> 0:08:46.880
<v Speaker 3>of what this can unleash for our society. Incredibly excited

0:08:46.920 --> 0:08:48.839
<v Speaker 3>to be part of it. But I don't think we're

0:08:48.840 --> 0:08:50.760
<v Speaker 3>part of a hype cycle at all. I think there's

0:08:50.800 --> 0:08:53.720
<v Speaker 3>a lot of innovation taking place, and you know, frankly,

0:08:53.760 --> 0:08:57.240
<v Speaker 3>the innovation that's taking place, any inventions that we're seeing,

0:08:57.679 --> 0:09:00.560
<v Speaker 3>it's just breathtaking. So no, I don't pally viewed as

0:09:00.559 --> 0:09:01.520
<v Speaker 3>a hype cycle at all.

0:09:02.240 --> 0:09:03.360
<v Speaker 2>Rene hassem Ce