OpenAI Unveils First Custom AI Chip With Broadcom

25:13Navy. So this amount, this up to$1 billion is just equity that they would raise. It doesn't include any debt that they may be discussing. Rebecca, you did a great job explaining what Hadrian does. It's basically an amazing factory or contract manufacturer. Generally, what is venture capital attitude right now to defense technology and this whole kind of re-industrializing America push? There's tons of excitement here, Ed. Investors in Hadron include Andreessen Horowitz through its American Dynamism Fund, Peter Schultz Founders Fund, Lux Capital. this effort to sort of reshore manufacturing and create this cycle where consumers and companies can just rely on production and consumption within the U.S.

26:00is of huge interest to these people. Bloomberg's Rebecca Torrance teaming up with me on that one, on Hadrian, potentially hitting $7.5 billion valuation. Thank you very much. Now, coming up later in the show, we're going to sit down with David George, Andreessen Horowitz, general partner on A16C's big bet on SpaceX, conversation that we're really looking forward to. Right now, this is what technology looks like in financial markets, particularly in equity markets. And we are all waiting for what is a big one. That is Micron reporting earnings after the closing bell. Right now, Nasdaq 100 modestly higher to flat underperformance in chip stocks and Bitcoin just shy of 61 ,000 US dollars per token.

26:53Welcome back to Bloomberg Tech. It's a big moment in time for memory. The biggest headline is probably that SK Hynix will try and raise$29 billion in a US listing. But the AI trade is going to get its next cue from Micron's earnings after the bell. There's a lot of literature on the Bloomberg about that. The stock's down about 1.3 % right now. It's highly analogous from what we saw in some of the earlier days of NVIDIA, 2023 to 2025. Massive growth year on year, both on the top and bottom line. But the market really wants a bullish signal. Not that the bottleneck that is memory is unwinding a little on the AI infrastructure build out, but the demand has staying power.

27:34So that is what we're watching for throughout the course of the day. Another top story, Elon Musk is reshaping his empire's balance sheet. SpaceX just pulled off a record$25 billion investment grade bond sale, taking on more debt while lowering overall borrowing costs across Musk's sprawling businesses. For more, Bloomberg senior private credit reporter Paula Sellison joins us. How did it go? It went very well. At the peak of demand. There were$89 billion in orders for what was ultimately a$25 billion bond deal. That allowed the company to tighten or get lower borrowing costs over the course of the marketing process.

28:11And investors are very eager to lend. There is a bit of a history lesson in this. I find it fascinating. So Musk takes Twitter private. X joins XAI. XAI merges with SpaceX, and then SpaceX does the biggest IPO in history. And if you look at how they kind of shuffled the deck with the proceeds of the bond sale and what they plan to do in paying down existing loans, it kind of takes us back to square one. Is that right? Well, this was really all one long, convoluted way of allowing XAI to become part of an investment-grade company and then be able to borrow at those borrowing costs. So, you know, we had the Twitter LBO, which ended up adding about $13 billion of debt.

28:52And then the XAI leveraged loan and high-eld bond sales, which added$5 billion of debt. So when SpaceX acquired XAI, there was$17.5 billion of debt with interest rates in the 9.5 % to 12.5 % range. All of that essentially just got refinanced initially through a$20 billion bridge loan that was then taken out with the investment-grade bond sale on Tuesday, and now interest rates around 5.5 % to 6.5%. We broke the story in just the days before the IPO that SpaceX had got IG rating from the three main agencies. That was really important, right? the story right now in the AI space is borrowing, and the IG rating has proved to be pretty handy to SpaceX.

29:34Absolutely. So there's so much need for capital investment. Companies just can't do it with cash flow. They have to borrow, and they're borrowing across all the different debt markets. But the best place to borrow for most of them is the investment-grade bond market, because it's very large, very deep pockets, and it's a cheaper borrowing cost. So being able to get those investment-grade ratings just opens up access to capital. That makes it so much easier to borrow all this money to finance the AI build-out. Thank you very much indeed. In the equity space, SpaceX actually higher again today,$158 per share.

30:08SpaceX, David George, Andreessen Horowitz, general partner, is on set with us right now. He led the firm's investment in SpaceX, a stake which is now valued at more than$10 billion and has proved to be Andreessen's largest return in history. And you're somebody that I've wanted to speak to about the company for a long time. There's a lot here, but I think just as an opening, reflect on the IPO, David. And I guess not just in a moment in time for Andreessen, but what you think it's signified for what is a very big and now quite diverse company. Yeah. Well, look, thanks for having me on, Ed. Great to be here with you.

30:43The IPO is just a milestone in the history of the company. I think it's a great event for the company to be able to access new investors, access capital that they couldn't otherwise access before. And one of the things that I'm most excited about as it relates to the IPO is the fact that retail was able to partake in such a big fashion. So if you look at the allocation that went to retail as a proportion of the overall IPO, it's around 30%. And so if you just take that portion that went to retail, it's almost the largest IPO by itself in the history of IPOs. So I think it's fantastic that retail can be a part of that.

31:21Why is that important, David? Sorry to interrupt you. Retail has proven to be an excellent group of buyers and holders of stock. So obviously, it's a large proportion of the ownership of Tesla, but they also are large owners of the Meg 7 at large. And so I think it's excellent for SpaceX to be able to access that. I think, you know, in large part, retail is long-term thinking, which aligns with the way Elon thinks and how he's built SpaceX. And so I love that they're a big part of it. Probably the question I get for you most commonly is about the lockup, what will happen? You know, I think a lot of the venture firms would say, well, we will redistribute to our LPs the stock and we expect lots of them to hold.

32:00But that would seem to make the retail holders important in what was like a very complex lockup. It wasn't straightforward. Yeah, look, I think this is a great step forward generally in how lockups are run. And I expect for IPOs going forward, this is probably going to be the way it's run. So Cerebris did something similar where you have the lockups come off gradually over time. And I think this is good. This is healthy. It allows the public market investors to be able to buy stock over time as opposed to at cliff events. For us, we're very long-term thinkers. We're happy shareholders of SpaceX.

32:35We think very much about the long-term. Obviously, what makes us so excited about the business is all of the things that can go right for SpaceX, all the optionality that you have. And so the way that we look at it is you start with a launch business. And in order to be successful as a space business, you have to reliably be able to get things to space. And so SpaceX has the infrastructure in an incredible way to be able to build applications on top. Obviously, Starlink is the first one. We expect that with Starship and V3, they've achieved some de-risking in terms of what they can bring up to space.

33:11Eventually, they'll have rapid reusability with multiple launches per day on Starship. it's a feat of physics magnificence that they can take something that's the size of larger than a football field send it up to space grab it with chopsticks and then reuse it again they'll rapidly be able to do that and that's going to enable all the applications that they want to build on top it also now has a name uh star mind yes as of last night and then you know elon musk engaging with others on X. There are many out there that believe that the timeline that was presented in the prospectus for Orbital Data Center, SpaceX said as early as 2028, actually we could see some pull forward on that.

33:54Again, linked to the tracking of Starship. Where do you sit in what's realistic near term? Yeah, look, I think the biggest hurdle to cross is just rapid reusability of Starship. And I think, you know, there's a consistent theme with Elon's companies and certainly in SpaceX, which is physics has been de-risked. And so now it becomes an execution question. And so as you think about timelines, we feel like it is inevitable that they can achieve these milestones. It's just a question of how quickly and at what cost. And so if you have Starship that can rapidly, you know, reliably get things up and back to space, we think that's the major unlock for bringing compute to space.

34:35I know earlier you were talking about the$850 billion of CapEx for AI data centers on terrestrial grounds. Right now, it's getting harder and harder to get data centers live on the ground here. I think it's a matter of time before we have it in space. I actually think I like to reframe it a little bit. I don't talk about it as orbital data centers. I talk about them as sort of airplane-sized GPU racks in space. So think of it as something like 72 GPUs up in space with big wings that are solar arrays that are kind of the size of a 737. And then you can have many, many of those in space. They've demonstrated that they can do this.

35:18They have 10 ,000 LEO satellites. So we feel like it's a matter of time and execution before they can do that. The physics has been de-risked. To the point about how it's getting harder and harder to do this on the ground, I think at a minimum, orbital data centers will be incremental capacity that you can have in space on top of what we have on Earth. And I think there's a case that in the fullness of time, the economics actually get better than on the ground. David, the firm Andreessen Horowitz has been involved with Musk companies in different ways for about six years, in different ways being XXAI, the different financial transactions that took place.

35:57For you, how much was this IPO a referendum, essentially, on Musk himself? Or how central to the fate of this company do you see Elon being? Yeah, look, Elon is the centerpiece of all of those investment theses that we had in backing his companies. You know, he's been remarkably strategic in how he's put the companies together. And, you know, he has done very well by shareholders. He's taken care of shareholders. He takes a lot of pride in that. and we appreciate that as his partner. I think as it relates to the go forward, he's been very smart. He and Brett Johnson and the team at SpaceX have been very, very smart about capital allocation.

36:38When they have done acquisitions, they have been remarkably strategic in terms of putting things together first and putting X with XAI, which is very logical. And then obviously the fit of XAI and SpaceX together is undeniable. So we expect that he'll continue to be very smart about capital allocation, take care of shareholders, and we appreciate that. Where do you stand, David, then on the transaction that everyone continues to talk about, which is a future where Tesla merges with SpaceX? Is it rational to your mind? I go back to the things that he's done in the past. And the things that he's done in the past is he's decided to do acquisitions of his companies or mergers of his companies when there is very strong strategic alignment and it makes business sense.

37:21So I wouldn't expect it to be any different going forward. There's this broader idea, if you extrapolate out from Elon Musk, that even at the growth stage, late private companies staying private for longer, late stage growth, that the investment thesis is still around the founder, the person at the top. I don't know whether you share that view of others, that even if you are talking in the tens, hundreds of billions of dollars of value, that value is assignable to an individual or a group of individuals? Yeah. Look, I think you can see reflections of this actually in the public markets. So I think you can see this in Elon companies, certainly.

37:59I think you could see this with Apple under Steve Jobs. I think you could see it with Meta under Mark Zuckerberg. But certainly in the private markets, this is a centerpiece of our investment theses when we're backing companies at any stage, at any size. I recently wrote a piece that basically said late-stage venture, which is our asset class, is not about capital markets. It's about founders. Late-stage venture is about late-stage founders and enabling them to think long-term, make big decisions, and make big bets. And oftentimes, they choose to do that in the private markets. And so our asset class that we play in, it's about$5 trillion in size.

38:40It's grown 10x over the last 10 years. It's larger than the Russell 2000. It's almost 20 % as large as the NASDAQ. So this has become a real asset class. And I think there's reasons that we should explore why it has evolved. Just give the asset class a title. What would you call it? Late stage venture, late stage privates. But where the value accrues to the individual. Yeah, where the value accrues. Yeah. And the centerpiece of the investment thesis is around the founder and the big decisions that the founder has to make. There's a lot of key man risk to model for. Yes, that's right. But I think the more important thing to model for is what can go right.

39:12Okay. You know, we always ask ourselves, you know, it's hard to imagine what could go right. And we even see this with mature companies. If you look back at the time when, you know, the iPhone came out, if you look at the 2009 consensus analyst estimates for what Apple was going to do and that in 2009 for the next four years, and then fast forward four years, they actually beat those numbers by 3x. And that's probably the most covered company in the world. So we like to think with a great founder, with a technology trend on its back, what can go right. And if you look at this last cycle that we've just gone through pre-AI, this cycle of mobile phones, social, e-commerce, SaaS, cloud all put together, the big story is that that big trend added about$25 to$30 trillion of new market cap.

40:00And I happen to think that on this AI wave, on the back of this AI wave, the trend is going to be even bigger. You head a growth team that manages$22 billion across five funds. But if you think about it holistically, you're basically in the top 15 private companies by valuation, which would have included SpaceX before it went public. How are you now deploying capital, new capital into new companies, you know, with those existing investments in mind? Yeah, look, our whole thesis is we want to be involved in the best companies at the earliest stage possible and then continue to back them every step of the way as they need capital.

40:39and often that has been through late stage rounds in the private market. Because you have a newer fund, right, which you can deploy capital out of. Yeah, we do. Yeah, we have our LSV5, Late Stage Venture Fund 5, as we call it. It's our fifth fund. It's about a$7 billion fund that we're deploying out of currently. And we see a ton of really interesting opportunities. So if you go back to the market that we're seeing and just look at our portfolio at A16Z and our latest fund, our portfolio dollar weighted is growing over 100 % year over year. If you compare that to the size, similar size companies in the public markets, those companies are growing about 20 % year over year.

41:19So we put a huge premium on very fast growth, great founders, market leadership, and sort of building on the back of the big trends. Also noting the cap tables look a bit different, like with the mutual funds, private growth equity. David George of Andreessen Horowitz, head of growth and general partner, I'm really grateful to have you on Bloomberg Tech. Thank you very much. Coming up, Cerebris reported its first ever quarterly earnings since going public. CEO Andrew Feldman's joining us. The stock is down 17%. There's a lot to unpack there. This is Bloomberg Tech.

42:00What if you could have more wins? More support. more sound effects. At LPL Financial, we like the sound of that because LPL offers more. Advisors, what if you could have more ways to help your clients? Ready to invest? What if you could find an advisor that really understands you? When it comes to your finances, your business, your future, at LPL, we ask, what if you could? Paid advertisement. Investing involves risk, including potential loss of principal. LPL Financial LLC member FINRA SIPC. Support for the show comes from public.com. If you're actively involved in your portfolio, you probably catch yourself repeating the same actions.

42:38Buying the dip, manually sweeping idle cash, putting on a hedge. On public, you can now create AI agents that handle all these tasks on your behalf. Just describe what you want to do in plain English. Like, if the VIX hits 25, buy a put option on the S &P 500. Or, if my cash balance goes above$20 ,000, move the excess into my direct index. You approve the workflow and your agent handles the risk. Monitoring the market, watching for your conditions, and executing your strategies exactly as defined. An investing platform driven by your intent, not just your clicks. You can also get full read and write access to your account via the public API.

43:18Go to public.com slash market and fund your account in five minutes or less. That's public.com slash market. Paid for by Public Investing. Brokerage services by Open to the Public Investing, Inc., Member FINRA and SIPC. Advisory services by Public Advisors, LLC, SEC Registered Advisor. Complete disclosures available at public.com slash disclosures. Whether you're planning a big tech event, launching a new campaign, or just stocking up on team gear, finding the right promotional products makes all the difference. 4imprint offers thousands of options, from on-trend apparel and premium drinkware to tech, totes, and giveaways, so you can find the right fit for any audience, purpose, or budget.

43:58You can customize it all. Your logo, your message, your look. And many items come with no setup charge to help you save. And if you're really watching the bottom line, you'll find standout choices at every price point so you can make a real impact while staying on budget. Plus, you'll get expert help, fast turnaround times, and their 360-degree guarantee. So you can be fore-imprint certain your order will arrive on time and look exactly right. Whatever your goal, 4imprint makes it easy to find your perfect promo match. Explore the possibilities today at 4imprint.com. 4imprint, for certain.

44:36Cerebrus reported quarterly earnings for the first time since going public last month. Its sales outlook beat Wall Street estimates, but still disappointed investors hoping to see the company carve out a bigger slice of the AI dentistry center market. Right now, shares down 17.5%. its biggest drop in its quite short history as a public company. What's behind that? CEO Andrew Feldman's with us. Welcome back to Bloomberg Tech, Andrew. You know, there was a time where we would talk about the merits of top-to-tail server ownership, how owning all of the content. Now we're going to talk about margin contraction, and we're going to talk about the stock being down 17%.

45:13That, to me, is kind of the mismatch. The outlook on the sales side beat Wall Street estimates. I think a lot of people are trying to understand the sequential margin decline. And for me, this is about ramping output for two big customers. Is that true? Yeah. I think what we did is we put forward a plan in the start of 26. We shared it with investors as we went public. And we're ahead of plan. We delivered record revenues of$191 million, up 92 % year over year. And for our cloud business, it was up 167 % year over year. We beat margin consensus substantially, and then we guided for full year that gross margins would be 10 % better than planned.

46:02We also shared that in Q2 and Q3, we would go back to some of our customers, and we would rent back gear that we'd sold them to try and keep up with demand. And that would have a margin impact on the order of 10 or 15 points. We did that to keep our customers close, to be sure we could keep up with their extraordinary demand for our product, for fast inference. And so that was the story. On every metric we put out, we're ahead of plan. Have the proceeds from the IPO actually allowed you to move more quickly in ramping up capacity? Yeah, I think capacity is the largest constraint right now for everyone.

46:51Data centers are. And we've significantly increased our ability and our pipeline for data centers, which is now very large. We announced a data center partnership with Bell Canada for 120 megawatts. That will be delivered in 2027. We are pursuing data centers across the U.S., in Canada, in Europe, in the Middle East. There are the vast resources that we have now at our disposal give us tremendous advantage in the pursuit of this, the limiting factor data centers. What you're talking about, like matter of factly, is buildings, not necessarily the compute, right? It's not what you guys are offering.

47:43how difficult right now is it to get moving in America or other markets to get planning approval, get the concrete, get the labor, get the thing built? That's the irony of this market, that the AI market is moving at blistering speed and we are being constrained by data centers which move with the speed of real estate. And so that is a problem that is being confronted by everybody in the category. by the hyperscalers, by the neoclouds, by the new generation clouds. Everybody is confronting this similar problem. Andrew, Cerebris does not rely on traditional off-chip HBM. Would you just explain that, the basics of the technology, but how insulated are you from the memory bottleneck that others are experiencing?

48:35Yeah, that's a really good point. Because of our innovative architecture, because of our wafer scale approach, which we don't use HBM. HBM is a type of DRAM, and it's made by three companies, one of whom is reporting shortly, right? That's Micron, Hynix, and Samsung. There's a global shortage. It's extremely expensive. Lead times are long, and we don't use it. So we have a tremendous advantage there. The other constraints in the supply chain for many are COOS, which is a process inside of TSMC. Again, we don't use it. And the third is capacity at the three nanometer node, that space in TSMC's factory that makes three nanometer chips.

49:22Again, we don't use it. We're at the five nanometer node. So our architecture has allowed us to deliver the fastest inference in the world by an order of magnitude while avoiding the main supply chain constraints faced by others in the field. can you say hand on heart, not just winning business, but have you actually been able to go to a customer and say, we can get this compute online faster than others for those reasons you just outlined, and then actually gone and done it? Oh, for sure. Any case studies? We signed, for example, we signed our contract with OpenAI on December 24th and had them in full production on February 1st.

50:11That's unheard of. That's quick. That's quick. Really quickly, Andrew, this morning, OpenAI is out with Jalapeno Intelligence Processor. But everyone is compute constrained, right? How do you interpret the Frontier Labs going to custom ASICs alongside their other compute options? Ed, one of the things we've been saying all along is that this market is enormous and is going to be met with a heterogeneous collection of architectures for hardware. This market will not be consolidated around GPUs. There will be ASICs. There will be ASICs from hyperscalers. There will be ASICs from labs. And then there'll be companies like Cerebrus with pioneering architectures who all of us will take big bites of this enormous market.

51:01I think one of the things that's difficult to get your head around is just how big the compute market is right now. That one of the things AI does is it makes tractable for compute much of the world around us. And that really wasn't the case prior to AI. Andrew, if the world is simple and it's money that greases the wheels to get your industry going, can we see you come back to capital markets in some form just so you can move quicker? We have more than$9 billion on the balance sheet, I think. So, we're really pleased with our position there. But of course, we are always scanning both the capital markets, both equity and debt for ways to accelerate our growth.

51:48Finally, how are you judging success yourself? What is the milestone that you'd pitch to the market to keep closest attention to? Well, I think if my mother's proud of me, I think that's the biggest thing you can ask for. I think for markets, I think when you lay out a plan that is aggressive and you crush it, uh that's how you feel good about both your ability to to execute and your ability to predict your own execution and so you know we we announced yesterday that we would beat our uh our full year we gave full year guidance that was 10 gross margin points above consensus um you ought to be proud of that and we got to continue to execute and continue to set extraordinarily high bars and then continue to beat them.

52:45Cerebra CEO Andrew Feldman, thank you very much indeed. Andrew alluded to it, Micron is the big one after the closing bell. Top line growth almost 300 % year on year. Net income growth 1 ,200 % year on year. The bar is very, very high. The market, it's very analogous to Nvidia, what we saw 2022 to 2025. Massive growth, but investors always wanting more. The stock's down 0.8 % going into that print. Another stock we're watching today, US fast food chain Wendy's surging as much as 42%. But wait, what does that have to do with tech? Well, it appears Wendy's has become the latest meme stock target.

53:26Shares shot up the rankings on stock twits, climbed to the top of the platform's trending list, seems to have been sparked by a now-deleted post on Reddit's Wall Street Bets forum that urge members to save Wendy's before it's too late. That does it for this edition of Bloomberg Tech. Recap everything on the podcast. You know where to find it. Online, Apple, Spotify, and iHeart, and all the Bloomberg platforms. This is Bloomberg Tech. When you're running a business, the best days are the ones where priorities stay on track. For midsize and large companies, risk can affect multiple parts of the organization at once, from property and liability to cyber and regulatory challenges.

54:02At that level, managing risk becomes an ongoing discipline. At the Hartford, the focus is on helping businesses manage risk before it turns into something more disruptive. And when losses do happen, that work is paired with insurance coverage shaped by years of underwriting, risk engineering, and claims experience. Learn more at thehartford.com slash risk mitigation. Policies provided by Hartford Fire Insurance Company and its property and casualty affiliates, Hartford, Connecticut. Coffee genius here. Most people see a busy cafe, but I see precision at every step. Thanks to genius from Global Payments.

54:38Transactions, instant. Inventory, precise. Operations, in sync. Absolutely genius. From sold-out crowds worldwide to managing the morning rush, genius keeps operations running smoothly. One Portado. Flawless pour, perfectly timed. Just beautiful. Big League reliability for any business. That's genius. These days, it seems like AI agents are just about everywhere you turn, every field and every function. But without identity, you can't trust they'll serve your business instead of jeopardizing it. Fortunately, Okta helps you get identity right by securing your AI agent's identities, giving you a single layer of control, a single standard of trust.

55:18So whether an AI agent supports a single user or your entire enterprise, with Okta, you'll turn risk into opportunity. Secure every agent. Secure any agent. Okta secures AI.