In short
Podcast Episode Summary: Bloomberg Tech - Investors Question Oracle’s Data Center Financing
Episode Overview Title: Investors Question Oracle’s Data Center Financing Hosts: Caroline Hyde and Ed Ludlow Description: The episode discusses concerns around Oracle's data center financing, OpenAI's potential $10 billion investment from Amazon, and Waymo's fundraising efforts.
Key Topics Discussed
Oracle's Data Center Financing
- Current Situation:
- Oracle's shares have seen significant losses, marking their worst day since December.
- Blue Owl Capital, Oracle's largest data center finance partner, has opted out of a $10 billion deal for a one-gigawatt facility in Michigan.
- Despite this setback, Oracle claims its data center project remains on schedule.
- Market Reactions:
- Investors express anxiety over financing difficulties, raising concerns about lenders’ perceptions of Oracle's financial health.
- There are substantial lease commitments (around $300 billion) that Oracle may be responsible for over the next 20 years.
- Comparative Analysis:
- Oracle's approach to data center financing differs from established players like Microsoft and Amazon, leading to market skepticism regarding its ability to transform successfully.
OpenAI's $10 Billion Deal with Amazon
- Funding Discussions:
- OpenAI is reportedly in talks to secure $10 billion from Amazon.
- The funds will be utilized to purchase compute power, specifically using Amazon's Tranium chips.
- Implications for Nvidia:
- The deal signifies a shift that could challenge Nvidia's dominance in the AI space. The demand for alternatives to Nvidia's chips is increasing as companies like Amazon and AMD enter the market.
Waymo's Fundraising Efforts
- Valuation Insights:
- Waymo is in discussions to raise over $15 billion, potentially leading to a valuation exceeding $100 billion.
- The company is recognized for its real-world applications with RoboTaxi services, providing a tangible product that may appeal to investors.
Investor Perspectives
- Portfolio Manager Insights:
- Tony Wong from T. Rowe Price expressed cautious optimism regarding the circular financing seen in deals between Amazon and OpenAI.
- He emphasized that as long as AI demand remains strong, companies like Oracle could present buying opportunities despite current market anxiety.
- Market Condition Reflections:
- There’s recognition that the AI market is not winner-takes-all; various companies can coexist and thrive as they target different segments of the AI ecosystem.
- The discussion also highlighted the cyclical nature of the market, particularly with memory makers like Micron expected to post strong earnings due to high demand.
Conclusion The episode delves into the complexities surrounding Oracle's financing of its data centers, alongside significant developments with OpenAI and Waymo. The discussions reflect broader trends in the tech industry, emphasizing both the challenges and opportunities amidst evolving market dynamics. Investors are advised to navigate these developments carefully, balancing risks against the potential for long-term growth in AI and cloud services.
Key Takeaways
- Oracle faces scrutiny over its ambitious data center financing amidst market skepticism.
- OpenAI's potential funding from Amazon signifies a strategic shift in the AI infrastructure landscape.
- Waymo's fundraising highlights the tangible advancements in autonomous vehicle technology.
- Investor insights suggest a cautious outlook on financing models, with opportunities still present in the tech sector.
For further details, listen to the full episode of Bloomberg Tech.
Written by AI. May contain mistakes. Listen to the episode to check what was said.
Transcript
Automatic transcript. May contain errors.0:00Hello, I'm Stephen Carroll. I'm in Brussels where many of Europe's biggest decisions get made. And I'm Caroline Hepker in London with the hosts of the Bloomberg Daybreak Europe podcast. We're up early every weekday keeping an eye on what's happening across Europe and around the world. We do it early so the news is fresh, not recycled and so you know what actually matters as the day gets going. From Brussels, I'm following the politics, policy and the people shaping the European Union right now. And from London, I'm looking at what all that means for markets, money and the wider economy. We've got reporters across Europe and around the globe feeding in as stories break So whether it's geopolitics, energy, tech or markets, you're hearing it while it happens It's smart, calm and to the point And it fits into your morning You can find new episodes of the Bloomberg Daybreak Europe podcast by 7am in Dublin or 8am in Brussels, Berlin and Paris On Apple, Spotify, YouTube or wherever you get your podcasts
1:02Bloomberg Audio Studios. Podcasts. Radio. News.
1:11Bloomberg 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. Oracle data center financing being questioned. Ellison's giant says the equity deal for its Michigan project is on schedule and doesn't include Blue Owl Capital. Plus, OpenAI is in talks to raise$10 billion from Amazon and will use its Tranium chips. We have the story. And Waymo is in talks to raise more than$15 billion at a valuation that could exceed$100 billion. Details later in the hour. But we begin with our top story. Oracle shares extending losses. Worst day that we've seen since December 10th, but the lowest since June of this year.
1:55Oracle's largest data center finance partner, that's Blue Owl Capital, will not back a$10 billion deal for a one gigawatt facility in Michigan. Still, Oracle says that the data center project remains on schedule and said its development partner, Related Digital, chose, quote, the best equity partner from a competitive group of options, which in this instance was not Blue Owl. For more, we turn to Bloomberg's Brody Ford covering all things Oracle. You've talked about how certain data center projects that are going to be there for OpenAI might be behind schedule. But have you heard, and are we reporting out, that financing is becoming more difficult to raise for these data centers?
2:32The big context on Oracle is that they are embarking on a historic build-out of data centers where it's really a logistical feat. It's quite a tight schedule they need to be on. And if it all goes right, it's transformational for Oracle and the entire industry. But all that needs to happen for that to get off is a couple things falling out, right? And that's the issue with what we're seeing today is that the FT is reporting that there's been some issues in getting the financing for one of their big centers. And that's giving investors a lot of anxiety that, hey, are these lenders seeing something that we are not?
3:08That's what's driving the fear this morning. What's interesting is you've been writing about the sheer scale of leases. And I think leases is something different because it's off balance sheet. And we're hearing it in a 10Q that they could have up to$300 billion worth thereabouts. in terms of their overall leases that eventually they're on the hook for paying. But how is this different? How is this sort of financing different from capital expenditure and the like? Right. So we're used to hearing about capital expenditures, which is I'm going out and buying computers, I'm buying wires, I'm buying the data centers themselves.
3:41But what's becoming more popular is renting the data centers, and that allows you to not pay everything up front and not have that be on your books directly. And so what we've seen is a big spike in these kind of future dated commitments that Oracle is the example that they have about$250 billion that they're on the hook for over the next 20 years or so. And most likely this Michigan data center would be one of them. It's just kind of, you know, one more pool of spending that we realize quite how large this build out is going to be. What's interesting, I think, is the Oracle is different from the other hyperscalers in that it's relatively new to this whole data center lease build out element.
4:23It used to be a software company. So is that something that also the market's trying to factor in? How good is it at this compared to Alphabet, Microsoft, Amazon, who've been in this game for a very long time? Yeah, that's a huge part. I mean, if Microsoft has 100 billion in commitments, people don't bat an eye as much because their cash flow is quite a different style than of Oracle's, right? I mean, Oracle's spending is such a large percentage of its overall company size because it's making a very ambitious transformation. Now, if you're a bull on Oracle, you say, yeah, well, you know, they are doing what it takes to transform their business.
4:57But a lot of things need to go exactly as planned for this to go right. You're going to be across how all of it does go right or as planned, really forward, across the Oracle story. Look, let's turn to, well, the company that's buying a lot of the data center or needs the data center, Compute from Oracle, and it's OpenAI. It's in initial discussions to raise at least$10 billion from Amazon. And we'll use some of the cloud giant's Tranium chips as well for more. Let's go to Bloomberg's Matt Day, who covers Amazon. Is this a circular deal? Oh, it sure looks like it. If you roll the clock back a month, OpenAI bought$38 billion of compute capacity from Amazon.
5:35That was a seven-year deal. It was primarily for NVIDIA chips. You know, per this news that emerged last night, it looks like Amazon may be giving some chunk of that back in investment to OpenAI. So$10 billion in equity so that OpenAI can continue to afford to buy compute, of which it's going to be buying from Amazon and its chips. But go to the chips part of this, because, Matt, this could be a real endorsement once again for Tranium, in the same way that Anthropic lent its name and its capacity for actually using them. Yeah, that's right. Amazon also a close partner of Anthropic. They've invested$8 billion in that company as part of a deal.
6:08that got Anthropic to use Amazon's own homegrown AI chip for their models, right? So this would definitely be a coup for Amazon's in-house chip-making effort, which analysts are still struggling a little bit to understand the effectiveness of to date, in part because there's just few enormous customers for this stuff. There's not a whole lot of big model trainers out there. So it goes without saying that if OpenAI can make the GPT suite work on top of Amazon's silicon, that's a big deal for AWS and for Amazon. Because it was such a big deal for Alphabet. and we think about the endorsement that its vertical model seems to be getting from the market, what does it mean for NVIDIA?
6:42Because as you mentioned, the previous cloud deal between OpenAI and Amazon actually was a ringing endorsement for NVIDIA chips. It means there's hunger for an alternative to NVIDIA. I mean, NVIDIA stuff works well, but it's expensive. It's in short supply, only one supplier. So you've definitely got increased credibility among the other folks who are putting out AI accelerators on the market. You know, AMD has an OpenAI deal, for instance. Amazon and Google's TPUs are all making a claim for some of that big NVIDIA business. All of this is about trying to finance the heaven-needing compute coming from the likes of OpenAI.
7:18Matt Day, brilliant reporting. Thank you very much indeed. Let's get the market reaction. Let's get an investor perspective for you. Tony Wong's with us. He is Portfolio Manager at T. Rowe Price's$12 billion Science and Technology Fund. You are named and own the who's who of the players within AI infrastructure and AI use, Tony. Are you worried? Let's start with Amazon and OpenAI. Are you worried about the circularity of deals once again at play here? Yeah, well, I think that whenever there's some type of creative financing, there's always some scrutiny that's on there. But I kind of view it as, you know, there is a synergistic partnership here and that, you know, OpenAI is growing rapidly.
7:58They need to compute. And it makes sense that they want to look for multi-source. And so I think going to Amazon does make sense. In addition, I think they want some of that upside. And so you're seeing, like, whenever there's a new frontier of technology, there's often companies that need to come together and build the ecosystem. And I think that this is an instance of this. And so I think it's definitely something to watch. But I think that what matters at the end of the day is that AI continues to progress. The scaling laws hold. The use of, you know, Chatsby T, LLens continue to grow. And I think that when you look at these tools that are coming out of the companies, they are like transformational in terms of how we are rethinking work.
8:43So I think what really matters is that the end demand continues to be really strong and that we're seeing progress on AI. I'm looking at your holdings. and the number one holding that you have at the moment is Alphabet. Interesting news with the new Gemini release this morning, which we'll get into. But NVIDIA is your second biggest. Now, are you worried about NVIDIA losing any sort of market share to Alphabet, to Amazon, to some of these other chip players? Yeah, well, I think the market is growing. And so, you know, this is an exponential growth curve. So I don't think that it's a winner take all.
9:17And so, you know, multiple companies can do well. And in some ways, I think when you step back and think about what's going on, you know, there is kind of a space race for AI, but they're going to multiple areas, multiple moons. And so they're all doing their own kind of domain expertise. I think like with the TPU, it's very optimized, you know, for a specific stack and for the Google ecosystem. And then I think NVIDIA is like kind of more the merchant, like broader platform, can run everything. And I think both can have a place. And you can see that Google's been doing really well over the last five years.
9:56So has NVIDIA, so has AMD, so has Broadcom. And so to me, I think that there's not a winner take all here. Okay, I love that sort of element of there are many races, maybe many moons, Tony, but talk to us about the moon shot that Oracle is making at the moment and the anxiety that people might not be wanting to lend to their future of leasing and the sheer scale of capacity that they're building with other partners, of course, out there like related digital. Have you got any anxiety around Oracle? I know it's in your portfolio last time that it was published. Yeah, so I think that, you know, Oracle definitely sees the demand signals that are probably really strong.
10:36And so they're looking to meet that demand, you know, in the near term with leases and then over the long term, probably owning their own capacity. And so to me, they see this as a big opportunity that is once in a generation. And so they're building for it and they're, you know, kind of being creative with it. And so to me, it definitely, you know, the market is super concerned. I think that there's probably somewhat of an overreaction. And I think as long as like, you know, the demand is there in years three to five, I think that that's what will make kind of the oracle that makes sense. So all of this anxiety writ large around sort of whether it's more experimental financing, creative financing, circular financing, is it actually a buying opportunity for you, Tony?
11:24I think we're constantly managing risk and, you know, rotating the portfolio to what has the highest ROI, what has the best risk adjusted return. You know, I still think that AI is like continue to be a really good area to place capital. as a lot of multi-year inflection. That being said, it's also been two, three years since ChatGPT has been launched. And so we are kind of further along in AI, but I still think we're pretty early. And so a lot of my views are expressed over the multi-year and think that you want to invest in companies that have strong competitive advantages that capture the value with their ecosystem.
12:05And I think that not everyone's going to be a winner in the space race, like some rockets might not make it. But to me, I think it still is like a really great overall theme to be invested in. Well, the moment you think the rockets to be betting on are in order of importance, Alphabet, NVIDIA, Broadcom, Apple, Microsoft. But how will that change as the application of AI comes? Thus far, it has all been about the AI infrastructure. But there's going to have to be proof in the pudding, the productivity. Are your science and tech fund going to end up looking like healthcare companies, like financial companies, like actually the users of this tech?
12:40Hey, I think that's a great question. And, you know, often I think about where the economic profit pools are shifting or where is the second derivative, like, you know, accelerating to. And you've been right. Like, you know, the last few years have been really heavy infrastructure buildouts. And that's where a lot of the companies have accrued a lot of value. You think about, you know, NVIDIA, Broadcom. Those have been my, you know, examples. And then you're also seeing the components that are going in the system like, you know, memory, HVM, you know, networking. Those companies are seeing a lot of like traction and economic profit accruing there.
13:20And then I think the next phase probably is on the application layer. And so, you know, there are large language models like, you know, ChatGBT, Anthropic, you know, XAI that are all working on this. And so, yeah, I think it's going to be exciting. And I think you might even see some previously viewed as kind of on the wrong side of AI, like perhaps be able to capture value. I think there's software companies that have kind of been written off here in enterprise that if you're a Fortune 500 company, often you're going to go to your software vendor once you've figured out what play you want to implement AI.
13:58So I think that there could be new companies that form on the application layer, But there could also be incumbents that are now leveraging AI, seeing the compute costs go down and, you know, the capabilities and models be more domain specific. So I think it's an exciting time to be in technology. It's constantly evolving. And I think that there will be also great opportunities, you know, that we don't think about right now. Well, you're in some of the software play, in particular Salesforce, which has been a bit beaten up for perhaps being not at the tip of the spear of things, AI, in terms of the monetization.
14:30But where it's really been monetizing has been the picks and shovels and memory of late. You just mentioned memory. We've got Micron after the bell. Are you feeling positive about how much more we're seeing price strength from these sorts of companies? Yeah, we're definitely on an upward trajectory in memory prices. So I think what's going on there is that HVM, the stuff that goes into these GPUs and AI systems, you know, they soak up a lot of wafer capacity. And so, as a result, it's creating shortages in broader DRAM. And then also, that's creating shortages in broader NAND. And so, you're seeing this ripple effect into other areas, like Sandisk, for example, has seen tremendous appreciation of their stock price.
15:14So, I think that there will be these areas where the economic profit can be distributed to even more that didn't participate as much over the last few years. What's really interesting, and correct me if I'm wrong, but I've gone to the BlueMug Terminal. and I've looked at the members that make up your fund, Lambda Inc., a private company, is in there. You, of course, at T. Rowe Price, know the opportunity there is in these crossover financings and the desire there is for private companies and exposure to it. So are you going to be seeing these companies going public? Are you looking more for private holdings as well as public?
15:46Yeah, thanks for asking about that. You know, at T. Rowe, I think one of our biggest competitive advantages is that we are both public and we're also one of the biggest financers in the private markets. And so we partnered up with Lambda. I think that they've got a really interesting cloud stack that is unique and they're on the right side of change and building these data centers really to what you would want a true AI data center to be like. So without a lot of the baggages that you would have to hold in traditional compute. And so I think the future is bright for them. The team's executing extremely well.
16:23and excited that they continue to progress in terms of timing on public markets. I think that they'll do it when they're ready. We'll keep an eye on Lambda and the executive team over there as well. Tony Wong, he's portfolio manager at T. Rowe Price Science and Technology Fund, public, private, we went everywhere. Meanwhile, coming up, the bidding battle for Warner Brothers is reaching a crossroads. That's up next. This is Bloomberg Tech.
16:59Warner Brothers Discovery. It is urging its shareholders to reject the Paramount Skydance takeover bid. Favouring its original agreement with Netflix, Bloomberg's Michelle Davis joins us now on the latest. We're seeing the market reaction. Paramount Skydance off by 4.5 % now. So what is the thesis of Warner Brothers Discovery as why the$30 Paramount is not living up to what's been offered elsewhere? So what Warner Brothers is saying is they reviewed the Paramount$30 a share offer. It's the same offer that Paramount had offered them two weeks ago when the board already deliberated and decided to go with Netflix.
17:34And in some ways they're saying this offer, the tender offer, is worse than what was initially offered. The logic is they don't view it as certain as the deal they already have signed with Netflix. And part of that has to do with the fact that Paramount has lined up, okay,$54 billion of debt, but they also are going to need$40 billion of equity. And as detailed in this 100-page filing that came out today, Warner Brothers' board says that they tried multiple times to get the Ellisons to make basically a personal commitment, personally backstop the equity commitment. But they, so Warner Brothers said, refused to do that.
18:11They're using a revocable trust to backstop the commitment, which Warner Brothers says is risky. You know, assets could be moved in and out of it. What they want is something that looks a little bit more like what Elon Musk did in the Twitter deal, which, if you remember, helped Twitter close that deal when Musk tried to back out of it. I mean, fascinating that everywhere in our show, it's questioning of Larry Ellison's financing for various things at the moment, whether it's his son's bid for Warner Brothers Discovery or whether it's his financing for future Oracle data center leases. It's interesting, though, also that others have been pulling back on the potential financing.
18:46Talk to us about what's happening with Jared Kushner's affinity. So yesterday, we reported that Jared Kushner's affinity was pulling out of the equity commitment or the backing of the Paramount bid. And this kind of underscored some of the concern that the Warner Brothers board has around where all the equity is coming from, because Paramount has said that it's lined up, you know, sovereign wealth funds across the Middle East, as well as affinity, as well as others to back the funding. And if those are falling through, what does that mean. It's also interesting because Kushner's involvement there had been seen as maybe something that would help Paramount clinch this if, you know, he is the ally to Trump that Paramount needs to get this through regulators.
19:24Without him there, I mean, it remains to be seen what that means. Warner Brothers has said that actually they view both the Paramount deal and the Netflix deal as on equal footing from a regulatory perspective. They say both can get through and we don't see any, marginal risk or difference between the two of them. And so it's kind of a moot point, I guess. We'll see how the formal rejection of Paramount's offer continues in this unfolding story. Quite the drama. Michelle Davis covering it all for us. We thank you. Now let's turn our attention back to large language models for a minute, because Google is rolling out a more efficient, more affordable version of its most powerful AI model across its products.
20:02The company announcing today the release of Gemini 3 Flash. And this comes just one month after the release of Gemini 3 Pro, which reasserted its leadership in the AI race. Since then, of course, OpenAI declared a, quote, code red and pushed out a new version of its flagship GPT-5 model, as well as an updated image generation model to cranky pace for Google AI's offerings.
20:28Tesla, it's facing a 30-day ban on car sales in California, the biggest US market. State regulators say the company's ads, well, they've misled consumers about its self-driving technology. Tesla's lawyers insisted that the ads were protected speech. Now, a judge backed the DMV's complaint. The automaker has 90 days to appeal or comply before the ban take effect. And Tesla has warned the ban could have major consequences for the business. Its shares have been on a tear before today. They hit your all-time high yesterday. Today, we just pulled back some, almost 3 % lower. Now, turning to other auto news, Waymo is in talks to raise more than$15 billion at a valuation that could exceed$100 billion.
21:07It's all according to sources. Now, the RoboTaxi maker has discussed raising billions in equity from external backers, as well as its parent company, Alphabet. Let's get the details. Simply beg Sarah Fry, who helped break this story. And really, we're starting to see the race being on$100 billion or in excess of is a lot more than what it was previously valued back in 2024. And really, those numbers, I mean, in the AI race, they seem equivalent to what we're seeing from, you know, OpenAI. There is a really big hope here for these companies. In Waymo, the difference is it has a product that is tangible.
21:46People are using it. People are taking their Waymos around San Francisco, right around, you know, see them outside my window all the time as regular transportation options. It is real. It is happening in many cities throughout the U.S. right now. And I think that that is why they need so much money. This is a very expensive business. This is a business that's going to require a lot of capital to expand into into other cities and also other types of roads. Right. Like freeways, more rural areas. I don't know how how quickly that's going to roll out. But we are seeing some international expansion in the next few months as well.
22:29So a very big moment for Waymo and for Alphabet. Annual revenue run rate. Our reporting is above three hundred fifty million dollars. Tell us just very briefly the Alphabet relationship here. They are being able to go to external capital. I've gotten a lot of questions about this because people are like, wait, isn't Waymo part of Alphabet? But Alphabet is investing in them. What's going on here? So it's one of Alphabet's so-called other bets. That's the division that includes these. We previously thought of them as moonshots, companies like Verily, the life science company, Waymo. and Ruth Porat under her leadership has tried to make this a more financially safe and smart line of the business.
23:14She's encouraging spin out, she's encouraging independence and so Waymo is getting a lot of external support as well and the idea is, yes. I have to leave it there, Sarah Fryer on all the news to do with Waymo. Congratulations on the scoop. This is Bloomberg Tech.
23:41Welcome back to Bloomberg Tech. We check in on these markets that have some anxiety baked into them at the moment when it comes to the financing of the AI spend. Still, the question of bubble lingers. We're off by 1.2 % on the Nasdaq 100. We're looking at some of the biggest draws in terms of the overall points perspective. And look, we've got a sea of red. and videos are up by 3.7 % when you're worth$4 trillion. That matters to an overall benchmark. We perhaps got some competition coming if OpenAI does indeed start using Tranium over at Amazon, a key story for us today. Broadcom once again under pressure, this time by almost 5%.
24:14Remember, it had been sort of hard since its earnings. We see it now at 324. Micron has its earnings after the bell. We're up by 2.5 % ahead of that number, but we're expecting 50 % growth in revenue for this business as it's managing to really capitalize on memory and the price of, we'll dig into that in a minute. But for now, we focus in on what's happening with Oracle. We're down by 5%. Once again, there is concerns about how Oracle is financing its big pivot into the world of cloud computing, the data centers, the leases, and indeed whether Blue Owl is able to be able to financing the latest one over in what's happening in Michigan.
Read the full transcript
24:48Let's talk about it with Mandy Up Singh, Blue Meg Intelligence Global Head of Tech Research. And we are again questioning Oracle's capacity to get into this data leasing pivot. Are you worried about some of the intricacies? Look, the reporting from Bloomberg is that Oracle is still going ahead with the Michigan project. It's not being delayed from a financing perspective. It's just not blew out behind it. Yeah. And look, I think a lot of it is around what kind of open AI run rate will we see in 2028, 2029? because we are talking about build-out of data center infrastructure for things that are going to go live in 2028, 2029.
25:29And that's where investors who are funding this, do they have the patience to wait for revenues to show up three years out? I mean, in the case of Oracle, had the backlog not been skewed to OpenAI, I don't think there would be as much anxiety. But because OpenAI seems to be losing its lead when it comes to, you know, the model provider race, and now Gemini, Anthropic, XAI, they all seem to have caught up. And that's where the anxiety seems to be stemming from is, are you better off just, you know, focusing on one LLM provider when you know it's going to be a race where, you know, four or five providers are neck and neck?
26:09And if they diversify that exposure in terms of having someone else like XAI use Oracle capacity, then I think that may calm down that anxiety. But for now, I think that's one thing. The other thing is training versus inferencing. I mean, if Oracle is more exposed to training workloads, then you have to ask yourself how much ROI will OpenAI have with training the next version of their model, which is going to cost a lot more. But will it have the kind of returns that everyone expects in terms of the model intelligence and how much they're going to get out of that next training run? So no one wants to underwrite training workloads, which is why Microsoft refused that to begin with.
26:55And all that open AI business went to Oracle. And now I think market is putting a question mark around why they should be funding Oracle, because training is not something that everyone is keen to invest in right now. Because, I mean, training costs will plateau. And it's just a question of when, not if. And we don't know when it's going to happen. There's also the question of what chips you can use for training, and we're questioning NVIDIA's dominance at the same time. What's so interesting, though, is the more we peel away the layers of the onion, the more we understand what this data center financing looks like.
27:29And a lot of it, we all talked about capital expenditure from all these big cloud companies. But actually, we've got to look off balance sheet. And we've got to look to 10Q reports like with Oracle. How much is that being factored in by equity investors as well as debt investors? I mean, even Emetta is looking to finance stuff off balance sheet. So that's where, you know, all these companies don't want to take too much debt. In fact, part of the reason why Oracle's equity seems to be going down is everyone is thinking maybe they should be, you know, raising this money via equity. Why raise everything using debt?
28:02And so that's where, you know, they're going to be creative. And sometimes you have to course correct. I mean, if, look, the CDS spreads keep expanding and the market is not ready to fund it, maybe Oracle may have to bring down its ambitions in terms of, you know, how much they will be looking to build. Because all this is for future capacity. I mean, if you have capacity right now, you know the new clouds are doing well. CoreWeave has a nice backlog. And you're supply constrained. So the question is, will that demand pattern remain for the next three to five years? And if it does, then Oracle will be fine.
28:37It's just a question of, you know, how much training needs these model providers will have and how much are you going to make from the next version of the model training? Is it really worth the ROI in terms of spending 5x on the next training run when things seem to be closing down when it comes to all these frontier models? At the end, it comes down to who has the lead when it comes to models. And for a while, it was just open AI. Mandy, saying of Bloomberg Intelligence, the breakdown there. Let's focus in on some earnings after the bell. Micron scheduled to report first quarter, fiscal first quarter results.
29:11And analysts expect the chip maker to benefit from strong pricing trends. Let's get to Kim Forrest, Boca Capital Partners CIO. We're off ahead of these numbers, but a lot of optimism baked into Micron and memory makers right now. Absolutely. And the price shows it. And there is a pattern to Micron's earnings, right? that, oh, maybe a month before people, and it's probably in reaction to Micron's customers talking about what they're buying from Micron. The stock generally, and when things are good, this is a very cyclical stock, but when things are good, things are very, very good for Micron. And I think in the short term, things are going to be really, really good for Micron, But often the stock sells off right after just because of that buildup that has happened, you know, in the recent past.
30:02So I mean, here today, Kim, the stock's up 168 percent. So we've got a little bit of a dip into the earnings. It's kind of nothing in terms of its performance. But what are you looking for after the bell in terms of signals that the strength is there and that we don't keep questioning the AI bubble as we know it? So there are very few providers of the big product that AI wants, which is high bandwidth memory, DRAM. And there's only two other players in the market. Micron has had a nice lead. The other players are coming up. But it looks like for at least the foreseeable future, infinite demand is there for this product.
30:44And I'm teasing, okay? I don't really believe it's infinity, but it's a very high demand. So we would like to know what that's doing. And you can see how Micron is shifting away from its older products and its more consumer-related NAND devices. They announced, you know, one of their longtime franchises, they're going to be closing that in favor of putting all its effort behind DRAM. So that's what I'm looking for is what is the why do they have that confidence? And why should we as shareholders, you know, go along with the ride? Because right now there is shaky confidence around certain names about the AI infrastructure build out.
31:25Do you have shaky confidence about certain newer players to cloud, newer areas and data centers? Well, this is going to sound like a Miss America kind of speech. I have a real, I know AI is going to work out. I just don't know how. The large language models seem to be consumers of tons of bandwidth, which you talked about on your last segment. Or not, you know, computing power, not bandwidth. But I don't know that that goes on forever because this whole scaling thing where just throw more compute power at it seems like the brute force method. And it's not very computer science. We like to think.
32:05I'm an ex-practitioner. We like to think rather than just, you know, build. So I'm thinking that breakthroughs in the design of these things are going to help out and may make a lot of the demand that we assume is there for hardware not irrelevant, but not infinite. Well, to that point, we've seen what China has been forced to do because it can't just throw ever more compute at the situation, doesn't have access in terms of supply chain. And what about Micron's own ability to the global opportunity here? And they're exposed to China in some ways. Sure. I mean, it is a concern because we don't really want to allow them to sell the really good stuff into China because of military issues.
32:49And, well, just, you know, leadership. But we're a capitalist society, so some of that kind of has to get pushed aside. and we let companies work throughout the world. That is a concern. But right now, for the foreseeable future, let's say 18 months to 36 months, there doesn't look to be another breakthrough other than scaling. And it looks like, all things being equal, that scaling is the path forward, especially for large language models. So unless information comes to light tonight, it looks like the companies that are offering products to these big data centers are going to continue to have their products very much in demand.
33:38What's interesting is we're trying to understand how much NVIDIA is still in demand at the same time as an Amazon builds out its Tranium, at the same time as TPUs come to the fore at Alphabet, and indeed Broadcoms used for ASICs. What do you make of where, if they can all win? Well, I think they can all win because the training, training, test, and validation are parts of this process. And they can better use different hardware setups. So I think that there's a place for everyone. What you're trying to figure out is how much of a place for each one of these items. I think competition always makes it better for the end consumer.
34:19And I'm not talking about the people using LLMs. I'm talking about the people building LLMs now. So I am very heartened to see that the leadership of NVIDIA isn't overshadowing everybody else with an idea. I think it's also kind of interesting just from a performance standpoint that companies like Google and Amazon have been able to quickly come out with something that's very usable. and we all had the assumption maybe 18 months ago that that was, you know, the horse was out of the barn and over the hill and, you know, so far away. NVIDIA was the clear winner. So actually, I love it as a capitalist.
35:00Kim Forrest, of Poker Capital Partners. Capital's in the name. We appreciate it.
35:10FinTech startup Imprint Payments has reached a valuation of$1.2 billion. The company that helps retailers offer co-branding credit cards has brought in$150 million in a new funding round led by Coastal Ventures. CEO Dara Murphy joins us now for more. What are you using the$150 million for, Dara? Hey, Caroline. Great to see you. Thanks for having me. At Imprint, our bet has been the same since we started the company five years ago. We want to be the modern co-branded credit card and loyalty platform for the world's great brands. And we've raised this capital to allow us to double down on that strategy.
35:40Our bet was relatively simple when we started the company, that the world's biggest banks weren't building good enough experience for great brands to launch co-branded credit cards. And that's been true. And what you see is we've taken a silent experience that used to live in the bank's app and put it at the heart of our partner's experience. And so when you open our partner's app or you open their website, you can engage with the card. You can pay your bill. You can check your balance. And that puts you one tap, one click away from buying your next bag of groceries and booking your next flight.
36:12And for our partners, that has meant huge increases in lifetime value. It's why brands like Booking.com, Rakuten, Crate & Barrel are choosing imprint over legacy banks and why we continue to focus on this mission with the money we've raised. A mission that was actually incubated in Thrive Capital. They participate in this round, Ribbit Dig, Kleiner Perkins, Spice Capital and the like, Dara. But I've sort of got a bigger question of just how many of these credit cards am I going to have, or even a debit card that you're looking at now? Am I, as an individual, going to have a Booking.com one and a Rakuten one, and I'm going to have a Turkish Airlines?
36:45At what point do I become exhausted with all the offerings? Well, what you see is we're working with partners that demand a lot of share of wallet from their customers, right? If you think about Booking.com, they're winning against airlines, they're winning against hotel companies, and customers are going there to book the full trip experience. And so by partnering with Booking or by partnering with Rakuten that's taking such a large share of online commerce for customers, we end up being a meaningful share of each customer's wallet. And so you're able to give them rewards for a huge portion of their spend.
37:15We would agree that very small fractional brands probably don't need credit cards. But more and more, we partner with brands that are parallel winners in their categories. They're winning in all of e-com. They're winning in all of travel. And so we get to give customers way more value for a bigger portion of their wallet. And so more and more customers sign up for that benefit rather than thinking, I'm going to have 50 versions of a credit card. Now, and do the inevitable AI question, but I imagine an awful lot of this is about personalization, and understanding where my next offering or next purchase is coming from, making sure you're meeting me where I am at.
37:47So how much are you looking at AI? How much are you having to hire in AI? Yeah, we think about it as kind of two sides of the coin, right? One is the customer experience that you've mentioned and one is how we build the company. On customer experience, automation and AI let us personalize how much rewards you get, how we talk to the customer, how the experience even shapeshifts for the customer and more and more investing in that. So it feels radically different to anything that a bank has ever provided in the past for these brands or for these customers. The really interesting thing, and maybe it's not as apparent under the surface, is we get to rebuild what looks like a financial company or even a fintech in the time of automation and AI.
38:27And so over the last year, we've grown the business by almost 300%. We've grown headcount by 20%. At every turn, we get to choose, are we going to build a new team here? Are we going to invest in technology and automation so we don't have to hire a team of people we can actually scale the company. We're using technology, which is an amazing opportunity that I don't think anybody has ever had. Wow. So I imagine that$150 million doesn't actually go that much on talent. It goes more on the marketing of your product. What does it double down on in particular? Yeah, three things. One is we continue to grow, right?
39:00So we obviously continue to market to customers, but we're not burning all this capital. If you think about the market we're in, big brands can choose 100-year-old banks or they can choose imprint. If you choose a 100-year-old bank, you probably sleep really well at night. If you choose a relatively new company, maybe you worry a little more. And so we just want to be very well capitalized to make that choice easy. Bucket two is more financial products. We have the credit card today that serves many of the brands' customers. But depending on your financial life, you don't want a credit card. Maybe you want a debit card with more rewards.
39:30So more of those products. And then the last thing is, as we've talked about, how do we build a company using automation and AI and scale without headcount, which is an amazing opportunity relative to where we would have been 10 years ago. You see public company FinTech today with thousands and thousands of employees. We get to make a decision every time. Do we hire or do we build? And more and more we get to build. Dara Murphy, Imprint CEO. Fascinating talking to you. Congratulations on the funding round. Thank you. Now coming up, after burning through more than$1 billion, a space startup with turning to a tech veteran.
40:05More on Relativities, Say that again. Relativity Space's new CEO. It's Eric Schmidt. This is Blue Bag Tech.
40:31It's time now for Talking Tech. First up, China's PDD Holding. It's creator of e-commerce site Temu, of course. It's fired dozens of workers following a fistfight between employees and Chinese regulators. And that's according to sources. Two fights occurred during a government visit to investigate claims of fraudulent e-commerce deliveries. Plus, chip startup Mythic, well, it's raised$125 million to support its efforts to challenge NVIDIA. Take a look at some of the investors. Mythic's tech relies on analog chips with far less power consumption than the traditional brains of a computer. And sticking with chipmaker shares, China's MetaX jumped nearly 700 % on their first trading day.
41:11It's the latest outsized market debut by Chinese chipmaker as investors are really betting on the domestic firms that could become viable competitors to NVIDIA. Now let's talk about another startup, Relativity Space. It set out to revolutionize the rocket industry with 3D printing, but it burnt through more than a billion dollars and its big idea hasn't exactly planned out. Now the company is turning to former Google boss, Eric Schmidt, and an$800 million fresh funding round from him to turn around the company. Let's get more on this. Bloomberg Space reporter Lauren Grausch. So a new exec, same thesis, still 3D printing of rockets?
41:47No, unfortunately, that seems to have been changed. And now the company is much more focused on traditional manufacturing, which, as you said, it's a far cry from what they started as, which was supposed to be this SpaceX disruptor that was going to revolutionize manufacturing rockets with 3D printing. But what we found after months and months of reporting is that they were really struggling with the 3D printing technology for a while. And over time, they started to slowly incorporate more traditional manufacturing into the process, so much to the point that 3D printing is barely involved in the rocket making process anymore.
42:31Well, this sort of speaks, therefore, more to Eric Schmidt's commitment that we've seen of late of competition versus China and supply chain and defense, space tech. But have you heard exactly why he's decided to take the CEO role? The overwhelming theory and what he has also hinted at on social media is that he purchased relativity for a trend that we've actually been hearing out about a lot lately, putting data centers into space. So you might have heard that from SpaceX CEO Elon Musk or Blue Origin founder Jeff Bezos because they've also talked about putting data centers in space. And so it seems that Eric Schmidt also has that plan as well.
43:11There might be some other things that he's cooking up that he'd like to do with a rocket company. But so far, that seems to be the prevailing reason he bought this. And we're now looking at, well, his fortune,$52 billion. He has money to put to work into this. So this is his own bet on space. Yes. So he was actually keeping the company afloat for a little while. And then he is, so far from what we've learned, has invested a substantial portion of money into this company. And then we've also heard that the company is potentially fundraising again. So they're actually going out to high net worth investors to see if they want to be a part of the company.
43:51So it looks like they actually do want to grow and maybe have big plans. But yes, for a while now, he has been injecting a lot of cash into this endeavor. long rush for the latest space startup to keep an eye on thank you very much indeed now that does it for this edition of bloomberg tech do not forget to check out our podcast you can find it on the terminal as well as online on apple spotify and iheart we'll continue on the theme of oracle of financing of data centers throughout the shows from new york this is bloomberg tech
44:31een
From the publisher
Bloomberg’s Caroline Hyde discusses questions surrounding Oracle’s data-center financing. Plus, OpenAI is in talks to raise $10 billion from Amazon and plans to use its Trainium chip in a challenge to Nvidia. And Waymo is in talks to raise more than $15 billion at a valuation that could exceed $100 billion.
See omnystudio.com/listener for privacy information.

