Bloomberg Audio Studios, podcasts, radio, news. Bloomberg Tech is live from coast to coast with Caroline Hyde in New York and Ed Ludlow in San Francisco. This is Bloomberg Tech coming up. Microsoft, Meta, Google and Amazon all report earnings when the market closes. blockbuster AI spending. But is there blockbuster AI sales growth? Plus, we continue to beat the earnings drum with the CEO of SoFi, which just released its own results that conversation later this hour. And Elon Musk is set to continue testifying today in his suit against OpenAI and its co-founders, Sam Altman and Greg Brockman, over the startup's pivot from a charity to a for-profit business. First, we talk about the businesses that are publicly traded, Ed, and naturally some resilience coming from the Nasdaq 100. We're up five tenths percent. There is a wall of worry and anticipation today. We know that we have a Fed decision not to mention the inflation pressure of oil once again showing up today. There is, as we know, a standstill on the Strait of Hormuz between the US and Iran. However, all fixated on one thing and one thing only. It is earnings. Yeah. 4 p.m. Eastern time, 1 p.m. Pacific time. Some people, this calendar is scary. To others, it's deeply exciting. Four of the biggest technology companies with a big focus across all of them on capital expenditures, the spending on AI infrastructure in particular. But a lot of people now want to see something coming out the other side. Growth. For three of those names, Microsoft, Alphabet and Amazon, it's cloud growth. Simple, for meta, just overall revenues. Advertising made better because of AI, Caro. Yeah, what is the adoption of AI? What is the return on AI investment? We look at this with just a mere 80 seconds for the four tech juggernauts of Microsoft, Alphabet, Better and Amazon to release results that will decide maybe the fate of this month's $5 trillion rally that we've seen on the S&P. Let's go to Bloomberg's Carmen Reinecke to talk about the high stakes trade heading into the bell today. Boy, just talk us around the 80 seconds because we've never really seen these four behemoths all come at once. Yeah, it's true. So if everything goes according to plan and everyone reports around the same time that they did last quarter, we're going to get the four biggest hyperscalers reporting earnings in 80 seconds. That's going to be a flurry of excitement activity on our desk, especially. And what we're really looking for here is how investors are reacting to the numbers that come out right at when they report earnings at these headlines. Top of mind, obviously, are going to be capital expenditures and then how those are balancing out with revenue growth from AI and cloud growth for the companies that have large cloud businesses. And there's so much at stake here. I mean, these are some of the biggest stocks in the S&P 500. And options data is showing that there's $750 billion of market value sort of on the line tonight as investors react to price swings. Carmen, how the shares of these four companies react really matters for markets broadly. Talk me through this chart, right? The market bottomed in March. And what we're looking at, if you strip out NVIDIA, Alphabet, Microsoft, Amazon have all been huge factors in the rally we've seen in equity markets to this point. Yeah, I mean, the thing that's been so interesting this year is we've seen a lot of breadth, right, coming back into the market. But I think last month or this past month has shown us overall that tech stocks are still the most important part of the S&P. And that's really what drove us back to record highs. There's just so much money in these stocks. So, you know, when we have so many reporting at once, there's the huge potential for huge market moves because there's so much market value just represented in these companies. And it's not just the hyperscalers. And we have Qualcomm tonight as well. We'll get Broadcom in a few weeks. And then tomorrow we have Apple. So these are the biggest companies in the S&P 500. They're the most important names in the AI trade. And so they're very, very important. Bloomberg's Carmen Reinecke. Thank you very much. Going into tonight, it's still all about CapEx. The question is whether Meta and Amazon and Google will need to raise more net capital in order to support this level of demand by investing in CapEx. So I think that it's going to be viewed negatively unless we see some material inflection in their top line growth. Our next guest believes AI demand is now verifiable at every layer of the stack. Natalie Gallagher, Principal Economist and Director at Board, joins us now. Pretty simple equation, isn't it? Capital expenditures in, what's coming out the other side in terms of revenue growth? How do you see that playing out in our economy right now? This is a really challenging point in time for the economy because the global macroeconomic backdrop is one of increasing uncertainty. We have the Strait of Hormuz, which is still struggling to act as a global energy artery. We have a blockade that has escalated from having a bit of leakage to being fully structural. And we have the economic impacts from that hitting, right, at the time where we're getting these massive releases. So the backdrop is absolutely fascinating. The cost structure is inherently changing. And so there's going to be quite a bit of demand on these hyperscalers to continue to deliver. Natia, I want you to explain that a little bit further. Let's bring up what the capital expenditures for this year are of those four big names, right? They are at those levels because they would argue that's what's required to build the infrastructure. to meet the demand that's out there on AI. But with the Strait of Hormuz, we've learned about the supply chain. You're basically saying we're in a higher cost environment now. So that capital expenditure is also impacted by the higher cost base of what it's being spent on. Explain it. Yeah, absolutely. I mean, part of the cost basis is transitory, right? So if the Strait of Hormuz reopened tomorrow, some of that volatility in input prices and energy, it would come back down to more of an equilibrium. Now, part of it is absolutely structural, though, as well. And that's what we really have to account for in the changing cost side of the equation. We have LNG and helium production that is taken offline for months to years. We have an added policy layer as well with the sulfuric acid export ban from China going into effect May 1st. And we also have premiums on key logistical routes that have been repriced. And history tells us those premiums don't go back to where they were in terms of pre-conflict. And so when we talk about overall CapEx investment, the cost side has inherently changed. Even if the Strait of Hormuz reopens again tomorrow, we're not going back to February 2026 pricing. And yet demand, from what we understand, remains resolute. We can factor in some of the reporting that was around OpenAI yesterday and whether or not it's meeting its own internal expectations for adoption of its use and adoption of its products. But what we seem to hear time and time again is that Microsoft cannot keep up with the sheer demand for its Azure product and indeed for the compute. We're seeing bottlenecks in memory and storage. Natalie, do you think we'll see the productivity gains that we need to see to vindicate this level of capex? Yeah, it's been fascinating so far. We have seen a bit of productivity movement. We've also seen industries that are more immediately exposed to AI. We've seen movement there that absolutely implies we can continue to see the productivity gains we need to see. Now, a broader question really being raised top of mind for myself is with this new cost structure, can it enable the continued infrastructure build out to support at the utilization trends? And I'm hoping to get some answers in the forward guidance of the releases tonight. Yeah, Natalie, if we get forward guidance that still says, yep, we're sticking to these numbers, we're currently showing that almost $650 billion is what we're anticipating, just these four names, Microsoft, Meta, Alphabet, and Amazon, to spend on capital expenditure in their fiscal 2026. Now, where could that come into some sort of logjam? Because we're hearing from, when we spoke to Andy Jassy, it's a worry about power. When we think about Alphabet's exposure or ability to get his hands on memory, that is something that analysts are bringing up today. Where do you think this could be not put to work as it's needed to? You know, I'm going to be looking for softness in cohorts that are more exposed to the macroeconomic headwinds. So thinking Asia facing cloud deals, right, European power grid deals. We should see some hints that even if the forward guidance remains persistent, there should be some signals, right, that maybe that's not a completely fair bet. Natalie, we have a Fed decision today. I'm only just going to state that because it's not on our calendar, but it's on many other people's calendar. We'll put it to one side. The executives on the call tonight are going to argue that AI is real and it's doing very meaningful things in the real world for the economy. Do you see any evidence of that? Yes, absolutely. I think we're in the early stages, right? So what I see in the data is that AI is augmenting workflows rather than replacing employment right now. We do see productivity gains. Are we seeing productivity gains to the extent that validates the current narrative from these CEOs? Maybe, maybe not. When we start talking about the Federal Reserve, the hope is that these productivity gains can give us an environment where we're actually in a lower rate, longer term environment because of those productivity gains. So absolutely tying back to the Federal Reserve and what a rate sensitive sector wants is lower rates. Nassali Yalia, Principal Economist and Directorate Board. Thanks so much for your time today. Coming up, we're sticking with earnings. And after SoFi's release this morning, shares are under pressure, and then some. We're going to talk to the CEO on loan origination, on guidance, all of that next. This is Bloomberg Tech. Let's take a look at shares of SoFi right now. In fact, down the most in a year, we're off by 13%, so the biggest move in a year, despite the company reporting strong loan origination. Let's talk through all of this with the SoFi CEO, Anthony Noto. And look, there was real strength in the quarter that you reported. And analysts are saying that time and time again, particularly in loan origination across the board. But while many are saying, look, you uncharacteristically didn't raise your revenue outlook. What held you back from doing that? Yes, we had a record quarter. Actually saw accelerating revenue growth to 41 year growth and reported over a billion dollars of revenue Our view was we went into the year with a view that we have at least two rate cuts and that was one of the key assumptions to our outlook for revenue guidance And while we beat the quarter, we're not raising full-year guidance because we now expect no rate cuts, and that would be a more difficult environment to operate in than if we had two rate cuts. The business is performing incredibly well. I'm not sure there are many companies that are generating over $1 billion dollars of revenue with 41% growth year of year and 31% EBITDA margins. We like to look at the rule of 40, which is revenue growth plus EBITDA margins. And we've had more than a rule of 40 for 18 consecutive quarters, including this quarter, and we're still forecasting that. But we saw no reason to raise guidance in this environment given the uncertainty as it relates to markets and interest rates, as well as global issues with the Middle East and the pressure on oil and inflation generally. A lot of macro headwinds that you're navigating. What about, in particular, private credit, Anthony? Because I'm looking at the slowing volume in the loan platform business. You said it's not because of that, but still analysts are kind of worried about it. Yeah, I think generally when you don't raise guidance, you give people something to worry about. So we had record personal loan originations, $12.9 billion record student loan refinancing originations, and record home loans. In fact, our home loans business doubled year over year, as did our student loan refinancing business. The loan platform business is one where we produce loans for other partners. And that business was very strong. It wasn't as strong as it was prior quarters because we made the conscious decision to put more of the loans we originated on our balance sheet because we get capital to do it. Those loans will produce cash flow over the next three years as opposed to a loan platform business where we just generate revenue in this quarter. We're not seeing any issues with credit performance. It's been quite strong. The concerns about private credit are really not appropriate for our business. Our partners in the loan platform business are buying consumer unsecured loans from us, not financing companies with corporate debt. your goal is to kind of be this one stop stop shop particularly for the for the category of young and affluent right we talk about that a lot with you on the program you didn't call out the economy this quarter and forgive me if i missed it but in the call you didn't call out the economy and a lot of people are looking at how solid credit quality signals are your boroughs are holding up would you tell me a little bit about how you see the economy but particularly people that are young and people that are affluent? Sure. We're not seeing any economic headwinds for our consumer. As you mentioned, they're younger, they're mass affluent. We saw very strong trends in spending through our debit exchange, interchange, revenue. We also saw strong trends in investing or investing business doubled on a year-over-year basis. And we're also seeing strong trends in performance in our credit card. So as we think about the breadth of our businesses and the activity that we see. We're seeing no slowdown in consumer spending, consumers' ability to deliver the obligations they have on their debt or their desire for more products like invest. So no issues whatsoever with our underlying consumer or their profile. I think the concerns about AI creating unemployment among white collar workers is a lot of hype. We're not seeing in any of our businesses. You're a technology company, but you also have this goal of being a top 10 financial institution. Also, Bloomberg Intelligence writes, you have that goal. When you talk to the team internally, how do you hold yourself on milestones to get to that? When will you decide, Anthony, that, yeah, you know what, today we are a top 10 financial institution? Sure. To be a top 10 financial institution, it's measured by market cap. We have seen a great increase in our valuation over the last eight years that I've been here. When I arrived, our capitalization was private, but the value of the company on a common stock basis was in the $2 billion range. We're now well over $20 billion of market cap. And so we're on our way there. We've made great progress in that eight years. We've taken our member base from 650,000 members to $14.7 million this quarter. our revenue from about $250 million of revenue to last year over $3.5 billion and on our way to $5 billion in 2026. But ultimately, we'll measure it based on market capitalization. And I think we're in rarefied air, growing 41% year-over-year with over $1 billion of quarterly revenue. And as I mentioned, 30-plus percent EBITDA margin. So it's just a matter of when, not if. And we're looking at the 20-plus percent increase in the share price over the last year. But today, is down by $3 billion. Does that frustrate you, the day-to-day gyrations, if that's how you're measuring yourself in terms of making the top 10? I think it's just the reality of the investment markets. The markets do not like uncertainty. We're not raising our guidance for the full year. People think there's some degree of uncertainty. Well, the outlook for the market is different today than what it was when we gave guidance three months ago. If the interest rates do actually come down, we'll see a big pickup in our business that would cause us to be more bullish than we already are. It's not like we're growing a very slow rate even on our current guidance. Our current guidance calls for 30% plus revenue growth and 30% margins. There just aren't a lot of companies with a billion dollars of quarterly revenue growing 30% with 30% margins. So we're focused on two things, driving durable growth through product innovation and brand building, and then delivering great returns. And that's what we continue to do. And so I understand why people are concerned about the outlook because we're going to raise guidance, but the world is changing every day around us. We're executing incredibly well. Our goal is to generate escape velocity so that we're the winner that takes most in the industry. As you mentioned, we're an everything digital financial services app. That's in place. The brand building's in place. The execution's in place. We just have to keep delivering and everything else will take care of itself. Anthony Noto, SoFi CEO. Great to have you back on the show. Thank you very much. Now, coming up, Apple plans an AI overhaul for its photo editing features. We've got the details next. This is Bloomberg Tech. It's time now for Talking Tech. And first up, Moonpay has acquired Sodot. It's an Israeli crypto security startup to launch a new unit focused on institutional customers. Now, the new business will connect large traditional financial firms to a variety of crypto and blockchain services as part of what they call, quote, a unique inflection point in the maturation of the digital asset ecosystem. Plus, a group of junior bankers tired of grunt work built an AI tool to end the drudgery of chores like formatting slide decks. Their startup, Rogo, now valued at $2 billion, promises to automate the tedious data entry and long been a right of passage on Wall Street and help financial firms enter more markets. And Moonshot-focused VC firm Eclipse just hired Meta's former VP of Generative AI, Amir Frenkel, as its first ever chief AI officer. Now, fresh off a $1.3 billion fundraiser, Eclipse is tapping Frenkel to help founders align their technical ambitions with AI Ed. Apple's earnings aren't out until tomorrow, but the iPhone maker's AI ambitions will be in focus for investors, and we've got an update on that effort. According to sources, Apple is planning a major overhaul of the built-in photo editing feature powered by its Apple intelligence platform. Let's get the details. Bloomberg's Mark Gurman. I absolutely love this reporting because every time we go down to Cupertino for a new generation of iPhone launch and we read your reporting, for lots of people, the camera's capabilities on an iPhone are so important. But therefore, so is the capability of what you do with those photos taken with the camera. So you're telling us that within the iOS ecosystem, big changes are coming. Yeah, this year is a very big year for both the camera functionality and the photo app functionality on the iPhone. You know, first things first, the iPhone 18 Pro and Pro Max are going to have a major camera update, one of the biggest camera hardware updates in the device's history using a new manual technology, a new manual sensor for letting more light in. It's a pretty cool concept, a cool idea that makes it more of like a pro point and shoot camera. On the software side, you're going to see a revamp to the editing features in the Photos app using Apple Intelligence. So right now you have the cleanup tool, which means you can, you know, take your finger, circle something and remove it from the image. It doesn't work particularly really well, but it is there. Three more AI features coming as part of iOS 27 that'll be announced in June. The first is the ability to reframe a shot. So if you shoot a picture in spatial, you're able to move the photo to change the perspective. So if you take a picture of a car and it defaults to the front of the car, you can tilt the photo so the perspective is perhaps the side of the car or even towards the back of the car. There's also an enhanced feature coming. You press the button and it uses AI to up the ante in the image, improve the color, improve the pop. And then there is expand, which is similar to a feature that Google has had for years on Pixel and Android, where you can drag the edges of the photo to increase the size of the photo. And then it'll use generative AI to fill in more content around the scene that you captured with the camera. Mark, go there. A feature that Google has had for years in Pixel. It does feel, again, kind of Apple's playing catch up here. Oh, absolutely. Apple is playing catch up. And it is a bit of a reversal. When they announced Apple Intelligence, they said they only had very limited functionality for AI related to the camera and photos because they believe in art. They believe in people taking pictures. They don't want to put AI at the center of that experience and take away from natural photography. Now, maybe that was their philosophy two years ago. I'd like to bet that that philosophy was only something they came up with because they didn't have those features. Clearly, AI is being implemented into every part of technology, into every smartphone operating system, and every part of the overall package. And so you're going to see them try to put AI everywhere that they theoretically can. Mark Gurman, fantastic reporting. Thank you very much indeed. Look, here's another AI application that's picking up steam. Humanoid robots, promising flexibility and the ability to learn and perform many tasks, But with billions of dollars invested, can these robots deliver real-world value when they fall short of the hype? In the last few years, there's been a lot of excitement about the potential of artificial intelligence, specifically with the breakthrough of ChatGPT which is a huge paradigm shift in the field of AI And so that the speculation What is the GPT moment for robotics Hear more from the team at Bloomberg Originals and our very own Ed Ludlow in it on today episode of Primer focused on humanoid robots. It's tonight, Bloomberg at 6 p.m. ET and on Bloomberg Originals at 8 p.m. Ed. The robots are coming. And coming up, we're going to come back on what to expect from big tech earnings. They are certainly coming after the market close today. More than $600 billion of CapEx commitments expected. Where is the growth and the other side of that equation when it comes to AI? And this is what markets look like going into it. We're sort of treading water at the Nasdaq 100 level. Chips outperforming, which has been a story for a while. Bitcoin, $76,200. Halftime, this is Bloomberg Tech. Welcome back to Bloomberg Tech. It's time for today's big number. And the big number is more than $600 billion. That's just across the four companies that are reporting earnings after the bell, the big tech companies. And it is the CapEx that they've either guided to for this year or that Wall Street has extrapolated out for the balance of the year. And we will keep looking at this, but it is an extraordinary calendar. After 1 p.m. Eastern today, Microsoft, Meta, Amazon, and Alphabet, the parent of Google, will report earnings within 80 seconds, 8-0 seconds. That's based on the timing of when they dropped last quarter. Brace, Caroline. Brace. Alphabet, let's start there. It's under the microscope today as investors look for proof that its massive $185 billion AI infrastructure bet is translating into real business games. Mandy Singh of Bloomberg Intelligence joins us with their preview and their analysis. You know, in some ways it's easy. We know the capex number there or thereabouts, so you can track how that's going quarter on quarter towards that goal. And then we have Google Cloud Growth. Is it as simple as that for Bloomberg Intelligence? No, I think it's more nuanced in the sense that Alphabet is one of the only companies out of the four hyperscalers you mentioned that actually gives their token metric. And what I mean by that is they have this monthly token trajectory around how much is the consumption of the Gemini model. And from that perspective, they are at least two to three times higher than the other frontier labs like Enthropic and OpenAI. And I think what I would want to know is how does that token metric translate into revenue growth? So it will show up in their search line. It will show up, obviously, in the cloud line. And, you know, the magnitude of the beat will kind of show at what rate are they monetizing their tokens generated versus what we know about Anthropic over the past two, three months and open AI. And from that perspective, I think that token consumption metric, for me, that is the key, along with cloud growth and cloud margins. But you want to see more of that tokens translating into revenue growth for Alphabet. So efficiency. Let's talk about GCP growth. cloud in particular. What could constrain that? There's some notes that landed on my desk today saying flash storage could be a constraint there for Google in particular. Is that a worry for you? Are they bottlenecks? Well, I mean, Google, like everyone else, relies on supply chain, you know, for a lot of the components from TSMC's fab capacity to, you know, getting memory from one of the three players to storage. So from that perspective, yes, there are a lot of constraints. And in this environment, if you haven't really prepaid for your capacity for all of these components, you will have a hard time sourcing, even if you are the size of Alphabet. So, look, all these things aside, I do think, you know, when Alphabet said their CapEx is going to be around $185 billion, they are, you know, putting the numbers and the, you know, investments in terms of getting that capacity. The question is, how much are they allocating for external use versus what is being allocated across their family of apps? And then external use, we will see that in the cloud line. But the internal use, you know, the AI overviews, the AI mode monetization, that will get reflected in the search and YouTube lines. And that's what you want to see in terms of ad pricing, because meta is expected to grow top line 30 percent. And ad pricing is quite visible over there. So you want to see that getting reflected in Google's results when it comes to search. Bloomberg Intelligence's Manip Singh, we thank you for pushing us ahead on Alphabet. Another stop we're watching today. Not earnings, it's Disney. Take a look at the two-day chart. Shares falling yesterday after the Federal Communications Commission said it would do an early review of the licenses of eight ABC TV stations. Let's get the latest of Bloomberg's telecoms reporter, Kelsey Griffiths. Now, the context of this is that there is uproar once again coming from the administration, directed in particular at one particular host on ABC. Can you talk us through it? Yes, Caroline. So we've seen a clash over the last year between the Trump administration and Jimmy Kimmel on ABC. What's interesting about the latest action from the FCC is that it doesn't mention Jimmy Kimmel at all. It actually says eight licenses that ABC holds in major markets are up for early review because they have potentially been engaging in discrimination internally. So this review is based on a letter that FCC Chairman Brendan Carr sent over a year ago to the Disney CEO investigating some of the hiring and employment practices inside of the company. So that's really what this review is about on paper. Internally, I am hearing that a lot of people do suspect that this is, in fact, related. Kelsey, there's an individual at the heart of this, which is Brendan Carr. Lucas was on the show yesterday, and we went through the kind of specific incident of the last week and then what happened at the end of last year. What role is the FCC chair? Does he play in this? What power does he or the actual committee actually have to do anything about it? The FCC chairman is all powerful in terms of what gets done by the agency. So Carr definitely knows about this probe. Nothing gets done without his knowledge and his consent. So that is one thing to keep in mind. The FCC would then kick off review of these licenses in a semi-judicial process. This could either play out in front of the full commission, where the commission essentially holds court and investigates whether the stations have been behaving in the public interest. And then it could also go to an administrative law judge internally at the FCC. Either way, the FCC chairman has a lot of oversight about how this process plays out. Bloomberg's Kelsey Griffin. Thank you very much indeed. Now coming up, the courthouse showdown between Elon Musk and OpenAI continues with another day of testimony from the Tesla boss we're going to discuss next. This is Bloomberg Tech. It is day three in the trial, putting some of the biggest AI players against each other. Elon Musk, you'll remember, is suing OpenAI, its co-founders, and Microsoft over the AI startups pivot from a charity to a not for a for-profit business. Let's get the latest Bloomberg's AI reporter, Rachel Metz. Rachel, look, Musk started his testimony yesterday. Remind us what we learned yesterday and what we can expect today. Sure. I mean, yesterday, it was a lot of talking through the founding of OpenAI, what happened in the early days, and Musk's feelings that this was a company that was created as a charity and that it cannot do this transaction that it has already done, which is to turn into a for-profit company. And we're going to see a lot of that. Right now, we're seeing a lot of that happening again today, a lot of talking to Musk. He's on the stand. He may be there through most or even all of today talking about right now the early days of OpenAI is what they're getting into, how the company was set up and why he says it was intended to be a charitable organization. We're watching pictures of Mr. Musk yesterday going through the metal detector several times, it seems. His principal kind of opening argument was that he needed to file this suit to stop Sam Altman looting OpenAI. Explain a little bit about what he meant and actually the core issues of this case. Sure. As OpenAI moves from being this company that had a pretty complicated structure but was chiefly considered a company that was a nonprofit to a company that has a for-profit entity, Musk is saying, hey, you can't do this. It was meant to be a nonprofit. So he's fighting against that. There's also a lot of other background and backstory there. There's been a long time acrimonious relationship between Musk and some of OpenAI's founders, chiefly Sam Altman and Greg Brockman, who Musk has filed the suit against. And Musk also has his own AI company that he wants to advance. So you've got a lot of competing opinions and ideas here about what should happen to OpenAI. OpenAI faces other lawsuits. Some of them are pretty emotive. One of them we understand regarding a mass shooting in British Columbia. Rachel, can you just talk us through what's being focused on there and what it means? Sure. So this morning, seven lawsuits were filed against OpenAI by families of victims of the Tumblr Ridge mass shooting in British Columbia. These are lawsuits filed by families on behalf of children who were killed in the shooting in February and also an educator and one child who has so far survived and is in ICU. What's at stake here is potentially OpenAI being considered responsible in some ways for the shootings. supposedly the alleged killer was using ChatGPT in the planning of the massacre and was kicked off of ChatGPT but then got back on and the company did not report it to authorities So that all a part of these lawsuits that have just been filed OpenAI issued us a statement for the story, which you can read in Rachel's reporting. They have a zero tolerance policy for using tools to assist in committing violence and that they've already improved their safeguards. I go read that and Rachel's reporting. Bloomberg's Rachel Metz, thank you. Now, sticking with OpenAI, AWS announced a partnership with the AI startup. a day after it ended its exclusivity deal with Microsoft. Amazon also plans to sell AI-powered productivity software for the office. We caught up with AWS CEO Matt Garman ahead of Amazon's earnings coming out later today. Well, we're quite excited about the partnership that we're announcing with them today. So, yes, so for a long time, when we built Amazon Bedrock as a way for our customers to access frontier models and access AI models, we've always started with the position that we wanted to offer choice, and we wanted to offer all of the best models available out there. And today we're excited to be bringing OpenAI into Bedrock. I think it's something that our customers have asked for for a really long time. And we're talking specifically about frontier models, not just open-weighted models, which was the extent of the relationship. That's right. We've had open-weight models for a while, but yes, this is the frontier model. So starting in preview today, we have OpenAI's model 5.4 and 5.5 is coming in the next couple of weeks. We're also collaborating on a new offering, which we called Managed Agents, featuring OpenAI, which is a complete managed agent capability so customers can really easily make stateful agents and build agentic applications together with Bedrock and OpenAI. And it's really just the start of a long-term partnership that we've established together. as the teams have gotten together, we see opportunity for us to really invent new capabilities for our customers to go and build interesting applications together. And I'm really excited about working together with the OpenAI team and unlocking more things that customers can build on AWS. We are in a strange situation where OpenAI models can now run on AWS, but Microsoft will benefit financially because of the latest terms of their agreement. agreement. They've ended exclusivity, but OpenAI continues to make payments to Microsoft. Yeah. How do you think about that? Oh, that's okay. Look, Microsoft has benefited from the growth of AWS since the very early days. In fact, we've supported Windows licenses as an example, I think since 2007, 2008, I can't remember the exact year, but so Microsoft has benefited. They build great software, they have good partnerships and they should benefit from those, but customers really want to use those technologies, SQL Server is a good example that runs great on AWS. In fact, many of our customers tell us that SQL Server runs better, way better on AWS than it does on Azure. And so for those customers, running in AWS is where they get the best reliability, the most security, and the broadest set of features. And so that's great. And if Microsoft benefits because their software or their partnerships or their licenses are being used inside of AWS, that's great. And in this case, we partner with Microsoft, just like we partner with Oracle and other providers that build software. And we want to make all capabilities available on AWS for people to build. The world of frenemies. It was a great interview, Ed. That was AWS CEO Matt Garman. Now coming up, more earnings. We'll discuss what to expect from Microsoft, Amazon, Meta. That's next. It's Bloomberg Tech. let's get back to tech earnings investors in amazon and microsoft expected to pay close attention to cloud capacity and cloud growth bloomberg's matt day covers both companies i don't know how you're going to do that after the closing bell let's start with amazon you and I spent time yesterday with Matt Garman, the AWS CEO. It is probably AWS where there will be a lot of focus. What are we bracing for? So Amazon Web Services is going to accelerate a little bit from the past quarter, looking like the fastest growth rate in three years. I think the big question is going to be when we start to see the fruits of all these new partnerships they've struck with OpenAI, as you mentioned. They're growing business with Anthropic. When does that sort of materialize? This cloud revenue is going to be one of the big questions this afternoon. People really felt that Amazon managed to get its ownership over its supply chain in many ways, and particularly when it comes to power, that's something that Andy Jassy has been speaking about for a long time. When you're looking at Microsoft, the fact that GitHub has paused co-pilot sign-ups, what does that reflect in terms of Microsoft's capacity, supply versus demand? So Microsoft has been struggling for more than a year now to get enough stuff up and running to power everything, both its commitments to OpenAI, its own software, its subsidiary GitHub, as you mentioned. And so they are just racing to put as much power and chips and data centers up as they can right now to kind of meet that demand. They're still growing at a pretty healthy clip cloud-wise, but it's certainly been clear for several quarters now they could be doing more if they'd had a better pipeline coming up. And what a share price drop we saw after their last quarter's earnings. We'll see if the 38% lives up to expectations today for Azure. Matt Day, thank you on all things Microsoft and Amazon. AI is also like to be a focus when we get Meta's results. Minna Smiley, senior analyst covering social media at eMarketer, is here to talk us through AI and plus some other things. There's a lot about the business model that's under scrutiny. But AI, CapEx, is that going to be where everyone looks? Oh, yeah. I mean, the funny thing about Meta these days is that that top line revenue growth number is actually like the least interesting thing about its earnings, right? I mean, we expect it to have growth once again. It's been posting double digit gains for years now because of its investments in AI that are helping it with better engagement, better advertiser targeting and stuff like that. But I think the real number people are paying attention to these days is that CapEx number. We do know that Meta has been spending exorbitant amounts of money on AI. I think that last we checked, it was $135 billion for the year was what they were expecting at most. So that's the number that we are seeing investors and onlookers really pay attention to and really try to get a sense of if there are going to be any changes or further comment on that spending. Comment also on how it helps fuel the flywheel effect to the business model, a business model that's been under some scrutiny in the courts. How do you think Mark Zuckerberg will react to that, some of the court losses that he's had in the United States? Yeah, it's a great question. I think people will really be hoping that they shed some sort of light on not only their reactions to these lawsuits, but sort of where they're going and how they're going to address some of this. because I think the reality is that, you know, with these lawsuits, it's still too early to really say, you know, how much meta is ultimately going to have to pay, you know, both literally and figuratively. I think, you know, we're still too early in the game. But there really is this sense that, like, something's got to give at some point, right? We're seeing this groundswell of lawmakers and parents and educators and, you know, all sorts of people really trying to argue that these meta and these other social platforms are dangerous for children. The fact that both juries found Meta liable does not look good for them. And so I think in the call today, it's going to be less about the numbers and more about how are they kind of viewing this sort of almost existential crisis to the business that could happen. I mean, it's hard to say to what extent it'll play out anytime soon, but three, five, ten years on the line, how are these issues going to play out and ultimately affect what these products look like? Minda, there are two big news stories about Meta that happened either in the quarter they're about to report or more recently. The release of MuseSpark as a model and then the layoffs that Bloomberg reported last week. How do those factor into this bigger conversation about how much Meta spends versus what they get as a result of that spending? Yeah, it's a great question. I mean, with Mew Spark, I think with the launch of their AI model, I think on the one hand, investors will be happy to kind of see, you know, the fruits of their labor, so to speak. We saw last year they spent so much money and time trying to, you know, kind of relaunch their AI business. And so, of course, having Mew Spark finally be here, getting a sense of, you know, what it looks like is definitely going to be comforting to some extent. But there are still so many questions around, you know, how is it going to make money? How is it going to compete with these more established players? So while it's something, I think it still leaves a lot of room for questions, certainly. And then, yeah, with the layoffs, I mean, that's another great question. I think there's been a lot of talk around, you know, to what extent is this AI washing? I also do wonder, you know, going back to the regulatory conversation we had earlier, I mean, could they perhaps be reorienting their business and trying to lay off certain people and hire new ones to kind of really address some of these scrutiny and these regulatory and legal concerns that could become a much bigger problem for them down the line. You know, Minda, it's so interesting. We're about four minutes into this conversation. None of the three of us have brought up the word advertising in the context of Meta. Why? Why not? Because at the end of the day, every quarter, it still comes down to advertising. Yeah, exactly. And it is a good question. It is wild to think that Meta, a company that makes most, if not almost all of its money from advertising. That's kind of why I think investors have been really kind of on edge about all of this CapEx spending. It's sort of like, at the end of the day, Meta is a social media business. So when you compare it to these AI competitors like OpenAI, like Anthropic, even Google, I mean, it's fundamentally different. And that certainly gives it some advantage, but it also brings about a lot of challenges. It's just, it really is a different beast. And compare it to the competition, YouTube, TikTok, where our eyeballs are going at the moment. Yeah. What are you seeing from the data? Yeah, I mean, our data shows, you know, by many accounts, meta is still, you know, I mean, we can see it in the revenue numbers, right? The fact that it's seeing double-digit growth, it's still, its platforms are still incredibly popular, namely Instagram. I mean, Instagram actually, by some accounts, is more popular than TikTok, at least here in the U.S. And so we are seeing time spent there, tick up among users. Facebook, I think people kind of often kind of like, you know, it's sort of an afterthought, but it's still a massive platform that a lot of people use. And we actually are seeing Meta kind of reinvest in it a little bit, try to make it more relevant again. There's a big push right now to get more creators on Facebook. Marketplace is still huge. So, I mean, yeah. I mean, of course, TikTok and YouTube, they are a huge platform that are doing very well. But like, yeah, Instagram, Facebook, WhatsApp, they're still quite popular. Minda Smiley, from eMarketer. Awesome. Thank you very much. looking ahead to crazy earnings afternoon. That does it for this edition of Bloomberg Tech. We're off to get a few coffees because we've got a few hours to go. The excitement doesn't stop. Don't forget to check out our podcast. You can find it on the terminal as well as online on Apple, Spotify and iHeart. Tune in after the closing bell. Thick and fast, this is Bloomberg Tech.
