WEBVTT - AI Spending Delivers Mixed Results to Stocks 0:00:02.520 --> 0:00:13.520 Bloomberg Audio Studios, podcasts, radio news. Bloomberg Tech is alive 0:00:13.560 --> 0:00:17.360 from coast to coast with Caroline Hyde in New York 0:00:17.640 --> 0:00:19.640 and Va Loow in San Francisco. 0:00:23.239 --> 0:00:26.360 This is Bloomberg Tech coming up. AI. Capex in focus 0:00:26.480 --> 0:00:31.000 after Meta, Microsoft and Tesla or report earnings with mixed 0:00:31.000 --> 0:00:31.960 fortunes FUS. 0:00:31.960 --> 0:00:34.400 Apple is also set to report earnings later today, with 0:00:34.440 --> 0:00:38.200 holiday season results and the impact of memory prices top 0:00:38.320 --> 0:00:40.240 of mind for investors. 0:00:40.680 --> 0:00:45.000 And Amazon reported hundreds of thousands of pieces of content 0:00:45.120 --> 0:00:49.120 it believed included child sexual abuse material, which if found 0:00:49.120 --> 0:00:52.400 in data gathered to improve its AI models. We have 0:00:52.840 --> 0:00:54.520 the Bloomberg reporting and it. 0:00:54.520 --> 0:00:57.520 Is significant in reporting at that, But let's turn our 0:00:57.560 --> 0:00:59.560 attention to what's happening on the markets right now, ed 0:01:00.000 --> 0:01:02.400 because it is a tough day, to say the least. 0:01:02.400 --> 0:01:04.360 It is all about capex in AI and you're going 0:01:04.400 --> 0:01:06.320 to go there with the individual names. But I point 0:01:06.360 --> 0:01:08.680 to a NASDAK one hundred that's having its worst day. 0:01:08.840 --> 0:01:10.399 Since November of last year. 0:01:10.520 --> 0:01:13.399 Two months now, we're down by two percentage points, some 0:01:13.600 --> 0:01:16.840 huge moves on individual names, but Crypto also languishing, were 0:01:16.920 --> 0:01:19.200 up by five percent. We're eighty four thousand at the moment, 0:01:19.400 --> 0:01:21.679 so much of being in digital gold in this currently 0:01:21.760 --> 0:01:25.919 debasement environment, it ain't working out for this particular area 0:01:26.040 --> 0:01:29.240 of gold because now it's just working as a risk asset. 0:01:29.280 --> 0:01:30.000 What are you looking at? 0:01:30.440 --> 0:01:32.360 Look, look, since we come on air like there is 0:01:32.360 --> 0:01:34.399 a bit of a vibe shift or a deepening of 0:01:34.480 --> 0:01:38.600 concern around AI capital expenditures. Two very similar stories that 0:01:38.640 --> 0:01:41.600 starts going in different directions. Meta told us this year 0:01:41.640 --> 0:01:44.000 one hundred and thirty five billion dollars of capex, but 0:01:44.200 --> 0:01:47.480 showed in the current period and beyond significant growth, particularly 0:01:47.480 --> 0:01:51.360 in the AD business. That justifies it. Microsoft capex one 0:01:51.400 --> 0:01:54.800 point five billion dollars beyond consensus in the court of gone, 0:01:55.000 --> 0:01:57.960 but thirty eight percent growth in Azure, the cloud unit, 0:01:58.040 --> 0:02:00.520 that's the one we look at for the AI probe. 0:02:01.040 --> 0:02:03.920 On a constant currency basis or currency adjusted basis, it's 0:02:03.920 --> 0:02:06.880 basically aligned with estimates. And people are worried that this 0:02:07.040 --> 0:02:10.840 growth in capex is outpacing what is now decelerating growth 0:02:11.040 --> 0:02:14.120 in Azure for the time being. Carried much to dig on. 0:02:14.080 --> 0:02:15.920 Here there is We've got the perfect person to do 0:02:15.960 --> 0:02:16.240 that with. 0:02:16.480 --> 0:02:19.920 Let's bring in Gabriella Borges Analysts covering the emerging software 0:02:19.960 --> 0:02:23.520 sector within Goldman Sex's Global Investment Research unit, maintaining and 0:02:23.520 --> 0:02:27.680 BI rating on Microsoft, citing long term strategic AI positioning. 0:02:28.040 --> 0:02:31.040 So do we have to swallow the sixty six percent 0:02:31.080 --> 0:02:34.160 increase in capex on the quarter for long term rewards 0:02:34.160 --> 0:02:36.200 with Azure that just aren't showing up quite yet? 0:02:36.800 --> 0:02:40.040 Yes, Hi, good morning. That's exactly the question that investors 0:02:40.040 --> 0:02:42.960 are asking us. What's really interesting is this is the 0:02:43.040 --> 0:02:46.120 second quarter in a row where Microsoft is actually telling 0:02:46.160 --> 0:02:48.600 you there is a more nuanced way to think about 0:02:48.600 --> 0:02:49.880 their capex allocations. 0:02:50.320 --> 0:02:51.480 Typically, what you would. 0:02:51.240 --> 0:02:53.760 Expect to see is capex goes up as your revenue 0:02:53.760 --> 0:02:56.640 goes up. There's a really direct correlation. What's happening in 0:02:56.639 --> 0:03:00.320 reality is Microsoft is telling you there are two importantrategic 0:03:00.360 --> 0:03:03.679 priorities in addition to Azure revenue. The first is Microsoft 0:03:03.720 --> 0:03:06.720 Copilot and what they're doing in the first party apps ecosystem. 0:03:06.919 --> 0:03:10.440 As Copilot adoption goes up, so does the capex allocation 0:03:10.480 --> 0:03:12.720 that has to go to support that business. The second 0:03:12.840 --> 0:03:14.840 is what they're doing with internal R and D. So 0:03:14.919 --> 0:03:18.960 think Microsoft AI for example, very strategic comparative because if 0:03:18.960 --> 0:03:22.000 they get some of these new markets right, like medical diagnostics, 0:03:22.000 --> 0:03:24.520 for example, that could be a really interesting long term 0:03:24.560 --> 0:03:25.800 product cycle for them as well. 0:03:26.160 --> 0:03:29.239 Amy Hood tried to get that new and it's across 0:03:29.280 --> 0:03:31.760 in the earnings call saying, look, we could have had 0:03:32.160 --> 0:03:36.480 marketly bigger percentage growth for Azure if we've served external clients, 0:03:36.520 --> 0:03:40.400 but we're serving ourselves with the compute. Why do we 0:03:40.520 --> 0:03:43.360 see the market overreacting like this even with. 0:03:43.360 --> 0:03:43.880 That new one. 0:03:44.000 --> 0:03:47.040 Yeah, it's a trade off between short term and long term. 0:03:47.280 --> 0:03:49.480 Short term you see that in the Azure revenue number 0:03:49.480 --> 0:03:52.920 every quarter. Longer term, though, is something like Copilot that's 0:03:52.920 --> 0:03:56.320 a gross margin accretion cycle and a long term monetization 0:03:56.440 --> 0:04:00.120 story that happens over multiple years. Microsoft AI. That's something 0:04:00.160 --> 0:04:02.080 that you're not You're going to get maybe a couple 0:04:02.080 --> 0:04:04.120 of blog posts. This here a couple of really exciting 0:04:04.200 --> 0:04:07.080 data points, but it's hard to predict one exactly you'll 0:04:07.080 --> 0:04:09.320 get those data points. So the way we see this 0:04:09.480 --> 0:04:12.200 is there's a little bit of a short term negative 0:04:12.200 --> 0:04:15.720 reaction here because it's hard to see into the future. Ultimately, 0:04:15.760 --> 0:04:17.640 we do still believe the data points will a croup 0:04:17.680 --> 0:04:20.080 positive on those two other strategic initiatives. 0:04:21.400 --> 0:04:23.840 Let's go back to basics a bit, Gabrielle and Karo. 0:04:24.120 --> 0:04:25.440 You know, we talked about this before the show. This 0:04:25.480 --> 0:04:27.760 is a drop of twelve percent, So this is the 0:04:27.800 --> 0:04:30.159 biggest drop in the stock since March of twenty twenty 0:04:30.560 --> 0:04:33.719 four hundred and twenty nine billion dollars of market value 0:04:33.760 --> 0:04:40.159 wiped off. That suggests serious concern. Gabriella, what do you 0:04:40.240 --> 0:04:44.720 think is the counter argument to that concern in the market. 0:04:45.600 --> 0:04:48.120 Yeah, I do think it's going to be a more 0:04:48.160 --> 0:04:52.080 balanced evolution on what enterprise AI adoption looks like this year. 0:04:52.520 --> 0:04:54.600 We know for two years now it's been a real 0:04:54.600 --> 0:04:57.760 struggle to see adoption tick up on the enterprise AI side. 0:04:57.839 --> 0:04:59.800 There's a lot of heavy lifting that has to happen 0:04:59.880 --> 0:05:03.200 on the data architectures, the data infrastructure, marrying some of 0:05:03.240 --> 0:05:06.839 these really cool llms with the richness of an enterprise context, 0:05:07.000 --> 0:05:09.440 and that's taken time. So as we go through twenty 0:05:09.480 --> 0:05:11.680 twenty six, one of the data points we'll be looking 0:05:11.720 --> 0:05:15.080 for is is that enterprise AI adoption starting to go higher? 0:05:15.279 --> 0:05:17.760 And what Microsoft told you last night is actually they 0:05:18.800 --> 0:05:22.200 have an increasing optic on adoption on the copilot side. 0:05:22.320 --> 0:05:24.479 We've seen data points on claud Cowork as well, just 0:05:24.520 --> 0:05:26.359 in the last couple of weeks. You've put all of 0:05:26.360 --> 0:05:28.640 that together with what Service Now is saying last night, 0:05:28.680 --> 0:05:31.400 and there is starting to be enough critical mass to 0:05:31.400 --> 0:05:33.920 suggest that enterprise AA adoption will go up from here. 0:05:35.279 --> 0:05:39.560 Net income was boosted by a gain in Microsoft's investment 0:05:39.600 --> 0:05:43.240 in open Ai, lifted pro per share earnings by a 0:05:43.320 --> 0:05:46.919 dollar and two cents, right, But our issue here is 0:05:46.960 --> 0:05:52.960 still how Microsoft manages the capacity issue that open ai 0:05:53.080 --> 0:05:56.440 presents them, as well as the circular financing issue. Right 0:05:56.480 --> 0:05:59.440 that you know that's the arrangement. Here is some capital, 0:05:59.760 --> 0:06:01.880 and here are the services that we'll charge you for 0:06:01.920 --> 0:06:05.160 as well. How do you think about that, Gabriella, and 0:06:05.200 --> 0:06:07.880 whether we now shift open ai a bit more into 0:06:07.960 --> 0:06:10.359 the risk column for Microsoft or not. 0:06:11.279 --> 0:06:15.039 So it's really interesting because Microsoft is taking steps to 0:06:15.240 --> 0:06:18.800 diversify their exposure to the AI ecosystem. Last quarter, we 0:06:18.839 --> 0:06:21.800 saw Anthropic come onto Microsoft Asia for the first time. 0:06:22.720 --> 0:06:25.000 We're also seeing Microsoft invest in their own what they 0:06:25.040 --> 0:06:29.040 call off frontier Microsoft AI models. Each of these steps 0:06:29.080 --> 0:06:33.080 diversifies the ways that Microsoft can win independent of whether 0:06:33.120 --> 0:06:35.240 open ai is doing well or not. With open Ai 0:06:35.279 --> 0:06:37.680 in particular, what we did see last night is that 0:06:37.720 --> 0:06:40.839 two hundred and fifty billion dollar contract that was announced 0:06:40.960 --> 0:06:43.240 about three months ago. Now you're now starting to see 0:06:43.240 --> 0:06:45.880 that in the backlog and that will roll on over time. 0:06:46.000 --> 0:06:48.840 The other positive catalyst to look for is the Fairwater 0:06:48.920 --> 0:06:51.200 Data centers. They're coming online over the course of a 0:06:51.320 --> 0:06:53.960 multia timeframe, but we'll start to get more visibility into 0:06:53.960 --> 0:06:55.560 that as we go through the next two quarters. 0:06:55.800 --> 0:06:58.040 We've had a bit of visibility about the silicon strategy, 0:06:58.040 --> 0:07:01.760 We've had visibility about the hardware side, but ultimately this 0:07:01.760 --> 0:07:04.200 company is getting caught up with software negativity. 0:07:05.279 --> 0:07:07.599 What about anthropic coming out? 0:07:07.720 --> 0:07:12.120 And yes, we see relationships, partnerships, but a cowork offering 0:07:12.280 --> 0:07:14.440 that could just not copilot out the water. 0:07:14.520 --> 0:07:16.320 How are you thinking about competition right now? 0:07:16.440 --> 0:07:20.560 Yeah, the burden of proof will be on Microsoft being 0:07:20.600 --> 0:07:23.440 able to show that the quality of outputs in a 0:07:23.560 --> 0:07:27.000 knowledge worker desktop and a knowledge work environment are better 0:07:27.480 --> 0:07:29.960 than what the competitors can provide. One of the key 0:07:30.000 --> 0:07:32.040 reasons they may be able to do that is because 0:07:32.040 --> 0:07:35.400 they're already entrenched in the Microsoft ecosystem. We already use 0:07:35.520 --> 0:07:41.160 Microsoft Excel, Outlook, Powerpoints, Poor Suite perhaps, and so if 0:07:41.200 --> 0:07:43.160 the intelligence that's in better in those apps can be 0:07:43.240 --> 0:07:45.600 married with the intelligence of the alms, you get to 0:07:45.600 --> 0:07:48.600 a better point from a productivity standpoint than if you bring. 0:07:48.440 --> 0:07:49.679 In at their party. 0:07:49.720 --> 0:07:53.520 Exactly. I'm so pleased we're going to this conversation. This 0:07:53.600 --> 0:07:56.560 has been so focused on earnings and like classic Bloomberg stuff, 0:07:56.920 --> 0:07:58.880 like I grew up in the generation of people that 0:07:59.240 --> 0:08:03.160 learned computed using Microsoft Word and Excel and PowerPoint, et cetera, 0:08:03.200 --> 0:08:06.720 et cetera. Irrespective of what I use today, I don't 0:08:06.800 --> 0:08:11.760 use that. I use Chat, GPT, Gemini, Claude. The totality 0:08:11.800 --> 0:08:15.800 and functionality of those platforms makes that kind of suite 0:08:15.800 --> 0:08:18.560 of software obsolete and I'm certainly not paying for it. 0:08:19.120 --> 0:08:22.400 How big a sort of existential question is that for Microsoft? 0:08:22.840 --> 0:08:25.600 And how reliant is it? Sorry to have a long question, 0:08:25.880 --> 0:08:28.640 How reliant on is it for copilot being a success? 0:08:28.680 --> 0:08:32.680 Therefore, Yeah, I think there are multiple ways to tackle 0:08:32.720 --> 0:08:35.839 this problem. I do think there is a disconnect between 0:08:35.880 --> 0:08:39.079 the consumer facing market and the enterprise facing market. Where 0:08:39.280 --> 0:08:41.480 we see this at our own organization. In order to 0:08:41.600 --> 0:08:44.920 get really good AI tools in the hands of people 0:08:44.960 --> 0:08:48.319 who work at enterprises, you need enterprise grade security or 0:08:48.320 --> 0:08:51.600 reliability depth the functionality. There are a number of systems 0:08:51.600 --> 0:08:54.400 that have to ultimately connect to build you something that 0:08:54.480 --> 0:08:57.120 is sustainable and durable over time. So I do think 0:08:57.160 --> 0:08:59.360 Copilot is one of the key ways that Masoft is 0:08:59.400 --> 0:09:03.000 addressing this problem. But there's also something like Foundry. Microsoft 0:09:03.000 --> 0:09:06.559 Foundry allows you to actually benchmark lms against each other 0:09:06.640 --> 0:09:08.680 and pick where you are on the frontier of cost 0:09:08.720 --> 0:09:12.280 and performance in a way that's quite differentiated today. And 0:09:12.320 --> 0:09:15.000 so there are multiple ways whereas at the infrastructre layer 0:09:15.000 --> 0:09:17.520 of the platform lay the application layer. Over time that 0:09:17.559 --> 0:09:19.720 will come together. And one of the things that we 0:09:19.760 --> 0:09:22.680 wanted to highlight today is just the strategic totalness and 0:09:22.760 --> 0:09:24.800 completeness of how Microsoft is thinking of us. 0:09:24.880 --> 0:09:27.839 Yeah, Gabriela Borg's of Goldman Saxe has been brilliant to 0:09:27.880 --> 0:09:30.200 have you on Bloomberg Text, thank you so much having me. 0:09:30.320 --> 0:09:32.800 Let's get over to Meta. Thank you. Shares of Meta 0:09:32.880 --> 0:09:35.240 right now on track for their biggest jump since July. 0:09:35.320 --> 0:09:37.760 The social media giant has e some of Wall Street's 0:09:37.760 --> 0:09:42.000 concerns about its massive AI spending after beating expectations in 0:09:42.040 --> 0:09:44.360 the holiday quarter, but it's all in the outlook. Joining 0:09:44.440 --> 0:09:47.959 us is Schweder Cajuria, Managing director of Global Internet at 0:09:48.000 --> 0:09:51.199 Wolf Research and outperform rating on the stock price target 0:09:51.280 --> 0:09:54.040 eight hundred and fifty dollars one hundred and thirty five 0:09:54.080 --> 0:09:57.240 billion dollars of capital expenditures at the top end, but 0:09:57.440 --> 0:10:01.079 also enough growth in the core B business aided by 0:10:01.120 --> 0:10:04.240 AI to justify it. Is it as simple as that? 0:10:05.760 --> 0:10:07.439 I think so. Well, first of fall, thanks for having 0:10:07.440 --> 0:10:11.880 me ed. I think the key question getting into the 0:10:11.920 --> 0:10:14.480 print was well, Meta is one of those companies that 0:10:14.559 --> 0:10:17.080 is spending a lot of money not only on AI 0:10:17.559 --> 0:10:20.320 headcount but also on capex, and they have yet to 0:10:20.400 --> 0:10:22.640 show something for it. Well, this is a quarter where 0:10:22.640 --> 0:10:25.880 they're showing it. Where the first quarter guide of over 0:10:25.920 --> 0:10:28.920 thirty percent year of a year on a reported basis 0:10:29.000 --> 0:10:32.160 is the highest we have seen since the third quarter 0:10:32.160 --> 0:10:34.400 of twenty twenty one and over four years and so. 0:10:34.720 --> 0:10:37.560 And at that time Meta's business was about half the 0:10:37.600 --> 0:10:40.200 size as it is now. So this is best in class, 0:10:40.280 --> 0:10:43.559 remarkable growth rate. So the point is if it can 0:10:43.600 --> 0:10:47.400 accelerate to this level, even if the growth decelerates going 0:10:47.440 --> 0:10:52.600 forward modestly, they have enough monetization levers that seem durable 0:10:53.160 --> 0:10:56.800 because they're making so many changes to their own models 0:10:56.840 --> 0:11:01.160 that are driving impressions growth and ro for advertisers in 0:11:01.200 --> 0:11:06.280 addition to potentially new monetization beyond advertising revenue that will 0:11:06.280 --> 0:11:11.559 come on because of their AI models over long near 0:11:11.600 --> 0:11:14.880 to midterm, and that is giving comfort on Okay, well, 0:11:14.880 --> 0:11:18.720 maybe they are spending to sustain a pretty healthy level 0:11:18.760 --> 0:11:19.800 off top line growth. 0:11:20.840 --> 0:11:23.960 Meta is not a hyperscaler, but something interesting. Caroline brought 0:11:24.040 --> 0:11:27.840 up with our last guest, Gabriella from Goldman Is Microsoft 0:11:27.840 --> 0:11:30.319 tried to talk up all the benefits it's feeling internally 0:11:30.600 --> 0:11:35.120 from allocating computes and investing. Susan Lee talks about exactly 0:11:35.200 --> 0:11:38.400 the same thing. Engineering output internally is up thirty percent. 0:11:39.400 --> 0:11:41.280 Is that a data point that you look at and say, Okay, 0:11:41.280 --> 0:11:43.920 I believe you that this is worth it because internally 0:11:44.120 --> 0:11:46.680 you're doing better as well. Yeah. 0:11:46.720 --> 0:11:49.280 Absolutely. One of the metrics that we do track is 0:11:49.320 --> 0:11:51.960 headcoune growth, and so what we are seeing is and 0:11:51.960 --> 0:11:55.120 what we expect is headcount growth to moderate growing forward, 0:11:55.240 --> 0:11:57.400 even though it will be in the positive territory, but 0:11:57.480 --> 0:12:01.319 more moderate as well as compensation skew to shift more 0:12:01.360 --> 0:12:05.960 towards engineering and higher compensated employees. In return, what we 0:12:06.080 --> 0:12:11.000 expect to see is perhaps greater level of productivity or 0:12:11.040 --> 0:12:15.199 shipments of products and new features that we think will 0:12:15.520 --> 0:12:20.280 then drive further ROI for advertisers and in exchange drive 0:12:20.360 --> 0:12:24.040 impressions growth, engagement, growth, and then monetization. That's how we 0:12:24.080 --> 0:12:25.959 see this play out, what. 0:12:26.040 --> 0:12:28.680 A difference the quarter makes and how much we beat 0:12:28.760 --> 0:12:31.400 up over CAPEX last time. In this time we see 0:12:31.400 --> 0:12:34.240 the optimism and the driving forward the promise of AI 0:12:34.360 --> 0:12:37.440 that understands us personally is how Mark Zakova was talking 0:12:37.480 --> 0:12:40.800 about it. But how crucial is therefore the next iteration 0:12:40.880 --> 0:12:43.280 of their large language model. How important is avocado, what 0:12:43.320 --> 0:12:45.200 the code name is or what they're building that it 0:12:45.360 --> 0:12:47.000 lands unlike Lama. 0:12:46.760 --> 0:12:50.839 Form, well, it is incredibly important. Going into the print, 0:12:50.880 --> 0:12:53.760 that was one of the key concerns for Meta that 0:12:53.800 --> 0:12:57.000 we haven't heard or seen a model yet when Gemini 0:12:57.080 --> 0:13:00.480 and arguably GPT five as well as Clouds models are 0:13:00.480 --> 0:13:04.760 way ahead. The expectation, however, is not for Meta's next 0:13:04.880 --> 0:13:08.520 model to really break the top three in the lam 0:13:08.559 --> 0:13:12.719 Arena rankings. It's more about Metas models to be in 0:13:12.760 --> 0:13:16.800 the ranking of in conversation of being a competitive model. 0:13:17.040 --> 0:13:19.880 If that happens, then that is a sign that they 0:13:19.920 --> 0:13:24.120 are on a good trajectory for upcoming updates to frontier 0:13:24.160 --> 0:13:28.679 models that they can leverage then for better impressions or monetization, 0:13:28.800 --> 0:13:31.600 and that should be enough to give investors some comfort. 0:13:32.160 --> 0:13:33.480 What excites you the most? 0:13:33.960 --> 0:13:36.920 Is it the commerce applications, the ways in which you're 0:13:36.920 --> 0:13:39.000 going to find it easier and easy to purchase using 0:13:39.400 --> 0:13:42.079 their own AI offerings, or is it actually that we're 0:13:42.080 --> 0:13:45.800 purchasing hardware through them. Billions of people can be wearing 0:13:45.840 --> 0:13:48.359 glasses and might be therefore wearing metaglasses. 0:13:49.360 --> 0:13:52.920 I think it's the former. They have shown a lot 0:13:52.960 --> 0:13:58.800 of competency in technology, in product development and then also 0:13:58.920 --> 0:14:01.840 in distribution, and so this is not a new thing 0:14:01.920 --> 0:14:05.640 for Meta. When we go back and Snapchat was picking 0:14:05.720 --> 0:14:08.599 up in their product, well, then we saw reels and 0:14:08.640 --> 0:14:11.600 it's over fifty billion dollars of revenue run rate. There's 0:14:12.160 --> 0:14:14.920 Open Ai, Sora, and guess what there is Meta ais 0:14:15.040 --> 0:14:19.000 Vibes as well. And they benefit from having a massive distribution, 0:14:19.520 --> 0:14:23.280 great resources, as well as the tech knowledge and product 0:14:23.360 --> 0:14:26.400 knowledge to scale these types of products and compete. So 0:14:26.440 --> 0:14:31.200 I'm more optimistic on their ability to drive monetization not 0:14:31.240 --> 0:14:35.240 only within advertising, because they are upgrading their models but 0:14:35.320 --> 0:14:38.880 also beyond advertising, whether to your point it's agentic commerce 0:14:38.920 --> 0:14:41.040 because guess what, a lot of gen Zers start their 0:14:41.080 --> 0:14:44.120 shopping and they start their search on social media. So 0:14:44.160 --> 0:14:47.680 there's a huge untapped opportunity there as well as business 0:14:47.720 --> 0:14:51.400 AI where they don't necessarily compete with anybody but themselves 0:14:51.640 --> 0:14:54.480 in getting a lot of those businesses on WhatsApp on 0:14:54.560 --> 0:14:58.440 board to spend on their business AI product that we 0:14:58.520 --> 0:15:00.840 think is a forty billion dollar opportunit from them. 0:15:00.800 --> 0:15:03.120 Shredo Cajuria, Always so great to have you on the show. 0:15:03.560 --> 0:15:05.280 Thank you from Wolf Research. 0:15:05.760 --> 0:15:09.600 Coming up, more earnings, more Capex to delve into. Tesla 0:15:09.600 --> 0:15:12.000 announces a new Terrafab build out as it looks to 0:15:12.040 --> 0:15:13.840 make its own chips. We're going to speak to within 0:15:13.920 --> 0:15:17.800 Tashakini of our invest as a blumber tech. 0:15:22.080 --> 0:15:26.960 I think in order to remove the constraint, the probable 0:15:27.000 --> 0:15:30.440 constraint in three or four years, we're going to have 0:15:30.600 --> 0:15:34.960 to build a Tesla terra fab, a very big fab 0:15:35.200 --> 0:15:39.280 that includes logic, memory and packaging. 0:15:41.120 --> 0:15:47.360 Domestically. That was Tesla's CEO Elon Musk outlining ambitious plans 0:15:47.400 --> 0:15:50.800 for a Terrafab, a chip factory for Tesla. In an 0:15:50.800 --> 0:15:54.600 effort to alleviate chips and parts constraints. Shares now down 0:15:54.640 --> 0:15:56.640 around two percent. Actually, at one point in the session 0:15:56.640 --> 0:15:58.520 we were up two percent and then as low as 0:15:58.800 --> 0:16:02.160 down four percent. The company's fourth quarter earnings top testimers 0:16:02.160 --> 0:16:05.800 everyone look past them. Tasha Kini, Ourk invest director of 0:16:05.840 --> 0:16:10.280 Research Autonomous Technology Robotics, joins us, Now, the kind of 0:16:10.320 --> 0:16:14.280 big headline for me was the capital expenditures commitment twenty 0:16:14.280 --> 0:16:16.840 billion dollars this year. You know full well, it's way 0:16:16.880 --> 0:16:20.200 beyond what they might do in a typical year. So 0:16:20.360 --> 0:16:22.880 let's start with that. You know, it's clearly a signal 0:16:22.920 --> 0:16:25.440 that these things that Elon Musk has kept you guys 0:16:25.480 --> 0:16:28.800 looking to the horizon on some time. Now it's on paper, 0:16:29.120 --> 0:16:30.040 it's time to act. 0:16:32.680 --> 0:16:36.760 Yes, well, you know, I think the capital expenditures are 0:16:36.800 --> 0:16:41.920 not too surprising to us. We had fairly aggressive investment 0:16:42.120 --> 0:16:46.200 estimations in our public valuation model. But that's all to say, 0:16:46.240 --> 0:16:48.360 you know, I think the next five years for Tesla 0:16:48.400 --> 0:16:51.360 will be dominated by the Robotaxi story. You know we've 0:16:51.560 --> 0:16:54.720 previously published. We think this could be attributable to over 0:16:54.840 --> 0:16:58.080 ninety percent of the enterprise value for the company in 0:16:58.120 --> 0:17:01.280 that time period. And you know, in order to put 0:17:01.400 --> 0:17:04.360 many vehicles on the road, which is their plan, they 0:17:04.400 --> 0:17:07.160 need to invest. Of course, a lot of that money 0:17:07.280 --> 0:17:12.399 is going into factories AI infrastructure build out. So you know, 0:17:12.480 --> 0:17:15.480 I think the story here is that Tesla, while they 0:17:15.520 --> 0:17:18.879 were you know, second or behind Weimo in terms of 0:17:18.880 --> 0:17:22.640 a commercial launch for ROBATAXI platform, they have an incredible 0:17:22.680 --> 0:17:25.359 scale that is really not matched by anyone in terms 0:17:25.359 --> 0:17:27.560 of the public announcements that they've made of putting actual 0:17:27.640 --> 0:17:28.840 vehicles on the road. 0:17:29.040 --> 0:17:32.359 Right on terror fabs specifically, you and the team at 0:17:32.400 --> 0:17:35.080 ARC know fool well, a fab at that scale is 0:17:35.080 --> 0:17:38.200 a fixed cost of twenty to thirty billion dollars, and 0:17:38.520 --> 0:17:41.480 the CFO was clear this is not reflected in their 0:17:41.480 --> 0:17:45.320 CAPEX forecast. But there's a reason that the industry doesn't 0:17:45.320 --> 0:17:49.000 put memory, logic and packaging on one site that doesn't exist. 0:17:49.119 --> 0:17:52.800 But it is Musk's proposal. How seriously are you taking that? 0:17:55.040 --> 0:17:57.600 Well, I think you know, he's always been a very 0:17:57.600 --> 0:18:00.800 ambitious CEO. I would interpret this as, Look, he's he's 0:18:00.960 --> 0:18:05.120 concerned Tesla needs a lot of chips. They have grand 0:18:05.160 --> 0:18:08.919 ambitions for AI data centers in space. We think you 0:18:08.960 --> 0:18:11.560 know that could uh that could be a really interesting play, 0:18:11.640 --> 0:18:14.440 especially as launch costs continue to decline. It could actually 0:18:14.440 --> 0:18:17.960 be cheaper than terrestri than for data centers on Earth. So, 0:18:18.920 --> 0:18:21.080 you know, I think the way to interpret it at 0:18:21.320 --> 0:18:24.160 is that he knows that he needs a lot of chips. 0:18:24.160 --> 0:18:27.919 We've seen him vertically integrate in the past in the 0:18:28.000 --> 0:18:30.720 electric vehicle business for instance, and parts of the supply 0:18:30.800 --> 0:18:32.520 chain that he did not see popping up in the 0:18:32.640 --> 0:18:36.080 US that they needed in the US. So you know, 0:18:36.240 --> 0:18:38.159 I think this sort of fits in with with what 0:18:38.200 --> 0:18:40.680 he's done in the past. Albeit, yes, it's a major 0:18:40.720 --> 0:18:43.840 announcement for Tesla, So i'd say broadly that, like I 0:18:43.880 --> 0:18:45.800 would interpret some of it as a call to action 0:18:46.000 --> 0:18:48.600 that you know, he he is looking out there at 0:18:48.600 --> 0:18:50.560 the supply chain. But yes, for now they have this 0:18:50.720 --> 0:18:55.840 very ambitious plan. And again I would focus on the 0:18:55.920 --> 0:18:59.480 at least from evaluation perspective, on the robotaxi picture here. 0:18:59.520 --> 0:19:02.320 Of course they also use those chips for optimists, but 0:19:02.680 --> 0:19:04.640 I would focus on robotaxis And. 0:19:04.640 --> 0:19:07.520 Let's focus just on what you have told the market. 0:19:07.560 --> 0:19:10.080 You anticipate if I'm right, by twenty twenty nine, what 0:19:10.160 --> 0:19:11.879 is it the stock will hit twenty six hundred and 0:19:11.880 --> 0:19:14.440 eighty eight percent of the company's value will be robotaxis, 0:19:14.720 --> 0:19:17.800 and you've talked about the millions of robotaxes that really could. 0:19:17.680 --> 0:19:18.680 Be produced by then. 0:19:19.680 --> 0:19:23.639 To that end, how integral is the XAI integration if 0:19:23.640 --> 0:19:26.639 we're going to think about integration, is that important for 0:19:26.680 --> 0:19:31.000 the story and the two billion well basically capital allocation 0:19:31.080 --> 0:19:32.040 that was made towards it. 0:19:34.160 --> 0:19:36.520 So you know, of course we hadn't previously accounted for 0:19:36.600 --> 0:19:38.360 that the way that I would think of it, as 0:19:38.520 --> 0:19:40.560 you know, on the call they mentioned, of course they're 0:19:40.600 --> 0:19:43.440 making GROC available in the vehicles. I think the more 0:19:43.440 --> 0:19:45.399 interesting parts of the story are, of course, you know, 0:19:45.400 --> 0:19:48.840 they could potentially share AI infrastructure investment. 0:19:48.480 --> 0:19:49.960 And mutually benefit from each other. 0:19:50.520 --> 0:19:53.439 He did mention that GROC could sort of act as 0:19:53.440 --> 0:19:57.560 this like maestro for optimists and for the robotaxis in 0:19:57.680 --> 0:20:01.000 terms of you know, like networking them together and sort 0:20:01.040 --> 0:20:04.399 of being the AI control center if you will, for 0:20:04.600 --> 0:20:07.600 robotaxis and any sort of like large scale optimist project. 0:20:08.240 --> 0:20:10.240 So that's how I would think about the AD investment. 0:20:10.760 --> 0:20:13.879 Let's talk about this ROBOTAXI unlock. Then what needs to 0:20:13.920 --> 0:20:17.199 fall into place for the twenty twenty nine moment to 0:20:17.240 --> 0:20:19.560 occur from your perspective, because at the moment, we've got 0:20:19.600 --> 0:20:23.040 just a handful in Austin that don't have someone in 0:20:23.080 --> 0:20:25.760 the driver's seat. How quickly does that scale do you 0:20:25.760 --> 0:20:27.119 think in realistic nature? 0:20:29.119 --> 0:20:31.600 Yeah, so you're correct that you know, there really are 0:20:31.720 --> 0:20:33.960 only a handful that currently have no one in the 0:20:34.040 --> 0:20:35.760 driver's seed. They did say though, that they have five 0:20:35.840 --> 0:20:40.200 hundred robotaxis and that they plan to double that every month. 0:20:40.240 --> 0:20:42.520 I mean, if that's true, if you look at Waymo's fleet, 0:20:42.720 --> 0:20:45.640 the last reported number we saw was about three thousand cars. 0:20:45.920 --> 0:20:49.400 So in other words, Tesla could surpass Waymos's robotaxi fleet 0:20:49.480 --> 0:20:50.479 within three months. 0:20:50.640 --> 0:20:52.439 I mean, I think that's incredible if you look at 0:20:52.560 --> 0:20:53.040 the fact that. 0:20:53.000 --> 0:20:56.439 Weimo launched the service back in twenty ninety, right, So 0:20:56.920 --> 0:20:58.840 again I would focus on the fact that this is 0:20:58.880 --> 0:21:01.919 like unmatched scale. They're a vertically integrated player. They do 0:21:02.000 --> 0:21:04.720 not have to partner with other auto manufacturers. This also 0:21:04.760 --> 0:21:07.320 gives them a cost advantage. So in the Big Ideas 0:21:07.359 --> 0:21:10.200 presentation that we just published, you know, we estimate that 0:21:10.240 --> 0:21:14.879