In short
Bloomberg Tech episode covering AI company growth and market implications, AI infrastructure financing, major AI/tech deals, and autonomous construction. Core segments include: Anthropic’s reported Q2 revenue jump and ARR run-rate; investors’ views on AI compute capex bottlenecks; Stripe’s planned acquisition of OpenRouter; Alphabet’s global debt issuance to fund AI spending; SpaceX family-office exposure ahead of IPO; China’s chip/memory momentum via CXMT; Bedrock Robotics launching the first fully autonomous excavator in the US; and Florida AI regulation politics.
Guests (and backgrounds)
- Todd Olson, Chief Investment Officer/Portfolio Manager at Parnassus Investments; 31 years in investing, focuses on infrastructure “second/third order winners.”
- Boris Soffman, co-founder and CEO of Bedrock Robotics; leads autonomous construction equipment rollout.
- Robert Shiffman, Bloomberg Intelligence Senior Credit Analyst; covers hyperscaler credit/debt.
- Natasha Mascarenas and Devin Pemberton and Peter Elstrom and others are reporters/interviewers (not guests).
Key claims
- Anthropic Q2 revenue reported at $11.5B, ~14x YoY; ARR run-rate crossed ~$47B in May; positive adjusted operating income.
- AI compute shortages drive capex and “wave of liquidity,” but investors should avoid over-leveraged circular financing.
- Stripe buying OpenRouter for >$7B to expand AI model access for developers.
- Alphabet raising ~$3.6B in Australia amid >$300B projected next-year capex; bond demand remains strong.
- Family offices held at least ~$3.8B in SpaceX (13F snapshot).
- Bedrock’s autonomous excavators are deployed on three paid commercial sites; uses cameras + LiDAR + on-machine compute (NVIDIA chips).
Notable examples
Sun Construction, Zachary, Champion Site Prep (Bedrock sites); Vulcan Materials, Linde, Hubble, GE Vernova (Parnassus examples); Salesforce/Workday/ServiceNow seat-license pressure (Olson); CXMT memory chips; Unitree Robotics IPO; Florida super PAC backers: Andreessen Horowitz/Greg Brockman vs Anthropic’s Dario Amodei.
Written by AI. May contain mistakes. Listen to the episode to check what was said.
Chapters
Tap a time to open that second in VOAnthropic's Revenue Surge
0:59 to 1:25
Discussion on Anthropic's impressive revenue growth and market impact.
“If you've ever waited on a refill or couldn't schedule an appointment, you get it.”
Anthropic's Revenue Surge
2:21 to 2:58
Discussion on Anthropic's impressive revenue growth and market impact.
“There is lots of technology news this Monday, but there are other things happening that are impacting the tech sector.”
Comparing AI Revenue Models
2:58 to 5:12
Analysts compare revenue models of Anthropic and OpenAI amidst IPO discussions.
“It's a 14 times increase from the same period one year ago.”
Market Dynamics and AI Investments
5:12 to 6:38
Exploring the market impacts of AI investments and liquidity waves.
“Where do we stand with Anthropic and OpenAI and sort of the cadence of them releasing the latest frontier models and where their business focuses are?”
Investment Strategies in AI
6:38 to 11:15
Insights on strategic investments amid AI growth and associated risks.
“Our next guest believes markets are still digesting the largest capex surge in history, driven by demand for AI compute.”
Risk Factors in Software Investments
11:15 to 14:00
Analyzing the risks in software investment models amid AI advancements.
“aggregates, materials, those physical parts of AI are going to be long bottlenecks in our opinion and we can actually underwrite the durability of those for a longer period.”
Grok with a Q's Funding Round
14:00 to 15:00
Learn about Grok with a Q's recent funding and future plans.
“Grok with a Q is raising$350 million in fresh funding, valuing the AI startup at about$3.5 billion.”
Stripe's Major AI Acquisition
15:00 to 17:00
Discover Stripe's acquisition of OpenRouter and its implications.
“I mean, this is at a time when the top conversation is how to make models cheaper, how to make the usage of AI cheaper.”
Thrive Capital's Investment Strategy
17:00 to 20:00
Explore Thrive Capital's approach to investing in high-value companies.
“Alphabet's tapping Australia's bond market for the first time, looking to raise about$3.6 billion dollars, the latest in a global borrowing spree as it ramps up spending on AI.”
Alphabet's Bond Market Move
20:00 to 23:20
Understand why Alphabet is tapping the bond market for funding.
“The people who seem to get more done than everyone else.”
Show all 21 chapters
Ultra-Wealthy Investing in SpaceX
23:20 to 26:10
Analyze the investment trends of family offices in SpaceX.
“Bloomberg's Devin Pemberton is here with the wealth beat.”
Tech Industry Updates: Anthropic, NVIDIA, and Okta
28:03 to 29:07
Discuss the growth of Anthropic, NVIDIA's investment in AI, and Okta's market performance.
“I need to probably use it more, but it just really shows the stunning pace of growth, Ed.”
China's Chip Industry Breakthrough
29:10 to 30:26
Explore China's rising chip industry led by CXMT and its market impact.
“China, a country that once trailed in hardware manufacturing, is experiencing a breakout moment for its chip industry, propelled by the success of CXMT.”
China's Semiconductor Strategy
30:27 to 32:12
Analyze China's strategy in semiconductor manufacturing and robotics.
“Apple, with this memory crunch that we've seen out there, wants to be able to buy chips not just from Samsung and SK Hynix and Micron, but also from CXMT.”
AI Models in China: Competition and Growth
32:15 to 33:20
Examine the competitive landscape of AI models in China and their implications.
“Of course, the government, the economy is very, very different from what we would see in the United States where there's much more government involvement.”
Defense Sector News: CEO Changes and Contracts
36:48 to 37:50
Update on defense industry developments including CEO changes and contracts.
“Let's get over to New York where Bloomberg's Yahaira Anand is standing by.”
Autonomous Excavators in Construction
37:51 to 41:21
Interview discussing the implementation of autonomous excavators in construction.
“Okay, for those who say the future will be built by robots, here's a site for you.”
Challenges and Future of Autonomous Construction
41:25 to 42:02
Delve into the challenges facing the construction industry and the future of autonomous equipment.
“What can you achieve in a 12-hour shift with an autonomous piece of equipment relative to a human-driven?”
Autonomous Construction Equipment Overview
42:02 to 46:26
Explore the potential and challenges of deploying autonomous construction equipment.
“Team, let's bring the pictures back up, because I think you can't get enough of showing it in action.”
Upcoming AI Regulation and Elections
46:26 to 46:49
Discuss the intersection of AI regulation and Florida's gubernatorial race.
“Okay, coming up, two groups with different ideas on how to regulate AI are backing the same candidate for Florida Governor.”
AI Regulation Candidates and Their Support
46:49 to 50:42
Examine the dynamics of AI regulation efforts and candidates in Florida.
“The CFTC is taking a closer look at one of AI's newest markets, tradable computing power.”
Transcript
Automatic transcript. May contain errors.0:00The most effective people at work aren't working harder than everyone else. They're working smarter, inside better systems. Superhuman Go, from the makers of Grammarly, is the AI assistant that works inside every tool you already use, always ready, already aware of what you're working on. It's a teammate whose only job is to help you be better at yours. With Go working with you, you can show off what you do best. See what Superhuman Go can do at superhuman.com. The thing about AI for business, it may not automatically fit the way your business works. At IBM, we've seen this firsthand. But by embedding AI across HR, IT, and procurement processes, we've reduced costs by millions, slash repetitive tasks, and freed thousands of hours for strategic work.
0:49Now we're helping companies get smarter by putting AI where it actually pays off, deep in the work that moves the business. Let's create smarter business. IBM. Healthcare doesn't always work great. If you've ever waited on a refill or couldn't schedule an appointment, you get it. That's the kind of stuff Optum is changing. They're using data and technology to integrate patient care, pharmacy, and everything else. So healthcare is connected, not complicated. What's that look like? Cheaper prescriptions that are easier to get and care that looks at the whole person. How you need it. Optum is helping make healthcare work as one for everyone.
1:25Learn more at business.optum.com. Bloomberg Audio Studios. Podcasts. Radio. News.
1:42Bloomberg Tech is live from the heart of Silicon Valley with Ed Ludlow in San Francisco.
1:51Ed Ludlow:This is Bloomberg Tech. Coming up, Anthropix second quarter revenue jumps to more than$11.5 billion, 14 times more compared to the same period a year ago. Plus, family offices around the world have built up billions of dollars of exposure to SpaceX. We'll have the details. And Bedrock Robotics launches the first fully autonomous excavator in the US. We're going to sit down with the CEO to discuss how the AI build-out is spurring demand for construction, and AI is answering the call. Good morning. There is lots of technology news this Monday, but there are other things happening that are impacting the tech sector.
2:27Ed Ludlow:The president has made two statements this morning. The first on Truth Social, that the absolute red line for the president is that Iran must not have a nuclear weapon. He also went on Fox News and said that if Oman gets in the United States' way, they will be subject to bombing. Both of those are impacting equities and Brent crude hovering around$89 a barrel. As I said, there is a lot of technology news. Let's take a look at today's big number. $11.5 billion. That's how much Anthropics reporting for their second quarter 2026 revenue based on documents seen by Bloomberg. It's a 14 times increase from the same period one year ago.
3:03Ed Ludlow:Let's get the latest. Bloomberg AI reporter Rachel Metz broke the story with the team. Let's get into all the numbers, right? These are data points presented in documents seen by Bloomberg and shown to prospective investors, right, in what is an interesting period of the year. There are several of them. Just take us through the numbers. Yeah, so we have this revenue number, which is many times what the company had reported just a year earlier, which indicates that what it is doing is trying to sell its services, especially to companies, but also consumers are buying its services as well before its chatbot cloud is working.
3:39I mean, it's up about, gosh, at least 14-fold from last time. It also is clear that the company had their annualized revenue run rate crossed$47 billion in May. From what we know, OpenAI has an annual run rate of over$40 billion, but it's important to remember that the companies calculate things differently.
4:06Ed Ludlow:I point out that there's a number of AI stocks on the move to the upside this morning. The other data point that's interesting is positive adjusted operating income, which in Anthropik's case, I guess, is making everyone feel good that losses won't be infinite and forever. But the comparison to OpenAI on ARR is really interesting because we put this in the context of two potential IPOs. If not this year, then maybe next year as well. What's the kind of reporting on the timeline for that? Oh, for me, it feels like it's still a little tricky to predict that one. But what I will say is that these two reports that we've done recently, this most recent one on Anthropic, and then the other recent one we had on OpenAI, and how their revenue is increasing.
4:51If we look at them individually and try to compare them to each other, because as I said, with ARR at least, they're calculating them in different ways. So it's certainly not an apples to apples comparison. It looks like each company is showing investors and potential investors better information than it could have just a month or two ago. And so that could be a really positive sign for an IPO.
5:15Ed Ludlow:What about the technology side? Where do we stand with Anthropic and OpenAI and sort of the cadence of them releasing the latest frontier models and where their business focuses are? Yeah, it's been kind of interesting to watch that unfold over like the last six months or so as these companies move closer to a potential IPO. You're seeing them sort of refocus or focus more intently on certain aspects of the business. I think with OpenAI, we're still seeing a cadence of releases, but we're seeing less of an array of things, which is interesting. It'll be even, I think, more clear as we go forward more this year.
5:56But we're also going to see the company continue to focus on things that it thinks are important, both on the consumer and the enterprise side, because those are both really big businesses for it. with Anthropic, I bet we can expect to see a lot more on the coding side, probably science side as well. We may see that out of both companies, both of those two things. But I think we're going to see a lot of that sort of stuff happening, this sort of coding, perhaps science, and maybe some other sort of work-related applications that we haven't seen too much of yet for these chatbots that they are hoping to sell to people.
6:30Ed Ludlow:Remember, it's Rachel Metz, who broke the story on Anthropic second quarter revenue Friday, but is now moving markets after that report this Monday. Thank you very much. Our next guest believes markets are still digesting the largest capex surge in history, driven by demand for AI compute. Todd Olson, Parnassus Investments, Chief Investment Officer and Portfolio Manager, is with us here in San Francisco on set. I want to start with the signal, I suppose, that comes from that report Rachel just gave. Anthropic has a real revenue number and an updated ARR number. and lots of different names are moving this morning.
7:05Ed Ludlow:Chip names, those that are invested in it, other software names. Why do you think that is? Well, thanks for having me, Ed. And after 31 years of being in the business, we're just at a stage now where the physical world just can't keep up with the digital world. That's creating incredible bottlenecks and that's widely known. And I think now you're seeing this wave of liquidity. And then we saw NVIDIA last week with the 500 billion announcement, Anthropix, SpaceX, OpenAI. So we're basically hitting the stage where a wave of liquidity is stoking demand into these shortages and bottlenecks. And that's creating incredible excitement.
7:40But for investors, we have to be careful to not get too swept up in that and then look at the second and third order winners down the road to make sure we're balanced and not getting ahead of our skis on beta in this investment landscape.
7:54Ed Ludlow:I've got questions about the NVIDIA news from last week. $500 billion where six Wall Street firms go and find third-party capital. And what I wrote about in my column this morning is there is a difference between depreciation and the economic life of a GPU. Now, as an investor, how do you model the economic life of a GPU and decide how to raise money against that? Well, clearly they're causing an asset class to be built and financed again with that wave of liquidity. You do believe that, that it can be an asset class of its own? I think it can, but one of the things will be the duration of the innovation.
8:28We're going to see incredible new architectures. We know that memory is in an incredible bottleneck right now. And so I believe that people are racing so hard today, it's hard to re-engineer these networks. But, Ed, we have to think about three, four, five years from now. We could be re-architecting and compressing memory and having new structures. What's the durability of these asset classes the next two to three years? I think it's very potentially in the bag. Five to 10 years, there's going to be tremendous change. And so I think we have to be a little careful on the risk and the adjusted return you're going to need to participate in these markets if you're a longer term investor.
9:09Ed Ludlow:Alphabet has looked to the equity market and it's looked to the bond market. And so far, everyone seems pretty sanguine about that. How do you feel about the activity, but also what you see as being investor demand in response? Yeah, it shows we're pushing into the more riskier part of this cycle. So initially, when the cash flow was all out of hyperscalers, you could capitalize that. Now we're going debt, equity markets. We talk about crowding out. We have a large national deficit. And we have homes to fund for the population. So we're getting into that more risky part of this infrastructure investment.
9:49It doesn't mean there's money to be made. It's just we're at a stage where it's riskier.
9:54Ed Ludlow:In the case study of the$500 billion with NVIDIA and the six U.S. investment firms, NVIDIA would say there's a degree of separation because it is those six firms channeling third-party capital, not NVIDIA's own capital. But there is still the circular financing debate. And for you as the CIO, where does that show up, that concern? Well, when we look at it, we want to make sure we're thinking ahead of the game and where we can participate as investors in the trend without taking on all that leverage and risk. So for us, in short, it's the second and third order winners. You think about a data center, it starts with a hole in the ground.
10:31So we're investing in Vulcan Materials, which is a toll booth on America's concrete. We own Lindy, which these purified natural gases. These are 15-year taker pay agreements. And so when you think about Lindy, they even help send rockets to space with industrial gases that are very pure, 15-year agreements. And we think about companies like Hubble and GE Vernova, that these things start with atoms to electrons to energy. So we're thinking about the more 5-, 10-, 15-year bottlenecks. And that's where I think as an investor, if you want to sleep well at night, you kind of look ahead where the infrastructure will go.
11:08and that's the long-term bankable asset class, energy, power, electrons, industrial gases, aggregates, materials, those physical parts of AI are going to be long bottlenecks in our opinion and we can actually underwrite the durability of those for a longer period.
11:25Ed Ludlow:When you look at NVIDIA, where there is a more direct circular financing concern is where it is investing its own capital into a project or campus where the party leasing the compute is a customer of NVIDIA's. But they're still investing in the same areas that you've just listed. Do you feel good about that or otherwise? We think Jensen's incredible. We love NVIDIA. We also like companies like AMD, which has great technology. Their Helios 450 RAMP is looking really promising. And they're basically showing up with a whole range of compute. and they're not as competing with their customers as much.
12:07So that's a company that we think is very well positioned in compute. If you want FPGAs, CPUs, which are much more important, GPUs and other embedded options. So that's another way to look at it where they're not as involved in the circular financing, but has incredible compute.
12:26Ed Ludlow:Todd, you've listed some very interesting names that are being presented with an opportunity in the build-out. What about those names that are most at risk of disruption, particularly in the software space? Yeah. So, you know, Salesforce, Workday, ServiceNow. These are good companies. We all have friends who work there. These have been great companies. But I think the seat license model is just going to be under a lot of pressure. You see the token usage. And the minute you see an anthropic revenue number, that's coming out of somewhere. And you see OpenAI. You're seeing Google Gemini. So I think the squeezing and the pricing out of that seat model is going to be very risky.
13:05And so we wish them well. They're good companies. I just think the long-term bankability of that is going to be challenging.
13:12Ed Ludlow:John. Todd Olson of Panas, System Investments on Set NSF. Thank you very much indeed. Another story that we're tracking in the markets is Cerebris. So it's up quite a lot. If you remember, we had Andrew Feldman CEO on the show last week where the stock had fallen quite a lot post-earnings. They immediately announced an expansion in more detail on their relationship with OpenAI. And you've seen momentum in the stock carry through Friday and now into Monday's session with it up 16 % this morning. We'll continue to track it. I do ask why Andrew who announced that 10 minutes after being on the show and didn't just say it during the conversation, but maybe one for another time.
13:46Ed Ludlow:Coming up, Stripe makes a major AI bet, clinching a deal to buy Open Router for more than$7 billion. We'll go to the private markets next. This is Bloomberg Tech.
14:07Ed Ludlow:Grok with a Q is raising$350 million in fresh funding, valuing the AI startup at about$3.5 billion. That's about half its peak valuation last year. The round set to be announced today was led by investment firm Disruptive, with NVIDIA also set to invest, according to a Grok representative. The company plans to use some of its recent financing to expand its total data center capacity to more than 200 megawatts by next year. This was a big one. Stripe is making a major AI bet, clinching a deal to buy Open Router for more than$7 billion, according to sources. The startup gives developers access to hundreds of AI models and could expand Stripe's foothold in the AI sector.
14:49Ed Ludlow:Bloomberg's Natasha Mascarenas. Who else part of the team that wrote the story? This was a big one. I saw lots of people share it in real time. You know, Stripe is big and this was quite a big deal for them. Yeah, I mean, this absolutely struck a chord. I mean, this is at a time when the top conversation is how to make models cheaper, how to make the usage of AI cheaper. And OpenRouter has, you know, founded in 2023, kind of soared as a result of helping developers access different models, optimize different models. And the Stripe buy is a sign of the value of that creation. I mean, a big guiding question for me, this entire AI boom has been who's going to capture the value and who's going to create it.
15:27And so very interesting and big news for open routers.
15:30Ed Ludlow:Very quick,$7 billion deal. But do we know how it's structured? Is this just Stripe flush with cash saying here's$7 billion? Yeah, I mean, I expect this to be a mix of cash and stock. and it's tracking for north of$7 billion. Last private valuation was about$1.3,$1.5 billion, so a pretty dramatic markup. You've been so busy for four days. On Friday, it started with you breaking the story of how one of Thrive's best funds has performed and some of the sort of firm-wide performance. You then had a follow on Saturday about them getting into Amazon, among many other stories. Just go through with the situation at Thrive.
16:07Yeah, I mean, the reason I was so excited to get the Thrive return numbers is really to answer the question about how much value is being captured in this era of concentration at all costs. It's this idea that Thrive, I would say really pioneered this idea of putting 90 % of the fund into your top dozen investments. A lot of people were like, that looks great if it's an open AI, but what does that look like when it happens across a bunch of different companies? And so the numbers show how much value they're sitting on top of. Of course, the next thing to break there is how much they'll actually distribute to investors, which is why that open router story is so interesting.
16:40It actually shows distribution.
16:41Ed Ludlow:It's super quick. Amazon is a public position. Yeah, so this tells us Thrive is basically expanding their public markets investment process. They want to stay focused, but I broke news that they're investing around 200 million into Amazon and that comes about two months after putting around 100 million into Shopify. Very interesting. Bloomberg's Natasha Masqueranis. Go on Bloomberg.com, check out all the reporting over the weekend. Alphabet's tapping Australia's bond market for the first time, looking to raise about$3.6 billion dollars, the latest in a global borrowing spree as it ramps up spending on AI.
17:12Ed Ludlow:But as big tech piles into debt, our investors started to push back for more. Bloomberg Intelligence Senior Credit Analyst Robert Shiffman is with us. Look, I asked you earlier this morning, how is this the same or different to what Alphabet's done in dollar denominated or other non-dollar denominated markets? And it's kind of a continuation of the same thing, right? Yeah, this is just a little blip on the radar screen. I mean, basically raising three and a half billion dollars in Australia doesn't really move the needle. It just shows that there's a considerable amount of demand around the globe for high-performing hyperscaler debt.
17:46Ed Ludlow:Why would a company as ginormous as Alphabet need to go to the debt markets? Is it just as simple as internal cash flows just can't sustain the wild capex commitment that's out there? Yeah. I mean, they teach you this day one in math camp. If you spend more than you make, you got to borrow. So that's what's happening here is they're just spending more than they're making right now. I think it's temporary. We continue to talk about a 2028 inflection. And I think these type of companies are going to become cash cows once again. The reality, though, is the more the media and all the smart money guys on podcasts and webinars keep talking about things like circular financing and how AI is not going to work.
18:29I think optimism continues to get stronger and stronger. And I think the bond market has proven that they are there to support this build-out. And in fact, if you just look at the last couple of weeks, spreads are meaningfully tighter, the widest names in our universe. Oracle, SpaceX are 30, 40, 50 tighter. Amazon did a 20, or sorry, Alphabet did a$25 billion bond deal just a week ago. Those bonds are tighter as well. So I just think people need to stop, you know, just listening to podcasts and start looking at the facts and that AI monetization is here. It's only getting bigger and the financing is in place to get it built.
19:04Ed Ludlow:So Alphabet kind of is top of the pile, right? They've done about$115 billion, excluding Australia, since the middle of last year. But you're basically saying investor demand is there and it will continue to be there. Investor demand is there. It's across the board. Remember, Alphabet is in the midst of issuing$85 billion of stock. They've done now about$75 billion of debt across currencies, Swiss francs, pounds, yens, Canadian dollars, already year to date. Listen, they're going to spend over$300 billion of CapEx next year. They're spending$200 billion this year. They need to fund it, but eventually we're going to see an inflection point.
19:41It's going to turn around. I think bondholders are going to win in spades.
Read the full transcript
19:45Ed Ludlow:Robert Shiffman of Bloomberg Intelligence, thank you very much indeed. Now, coming up, Elon Musk's SpaceX gets a boost from the world's ultra-wealthy. We'll tell you what recent 13F filings reveal about their holdings. That's next. This is Bloomberg Tech.
20:04The people who seem to get more done than everyone else. They're not working longer hours or running on more caffeine. They've just stopped wasting time on the stuff that doesn't move work forward. Switching apps, re-explaining context, hunting for files. Those aren't small inefficiencies. They're hours wasted every week. Superhuman Go gives you those hours back. From the makers of Grammarly, Go is an AI assistant that sits inside every tab and tool you already use, always available and already aware of what you're working on. Ask it to draft something, summarize a long thread, pull up a file, or prep you for a meeting.
20:45Go handles it without you ever leaving the page you're on. This is what it looks like when AI actually fits into your work, instead of adding to it. It's like having a teammate whose only job is to help you be better at yours. Go keeps up so you can move forward. With Go working with you, you can show off what you do best. Superhuman Go. Find out more at superhuman.com. That's superhuman.com. The thing about AI for business, it may not automatically fit the way your business works. At IBM, we've seen this firsthand. But by embedding AI across HR, IT, and procurement processes, we've reduced costs by millions, slash repetitive tasks, and freed thousands of hours for strategic work.
21:32Now we're helping companies get smarter by putting AI where it actually pays off, deep in the work that moves the business. Let's create smarter business, IBM. It's time to plan ahead and make sure your brand is showing up in ways that can have an impact. That's where 4imprint comes in. 4imprint have promotional products that work as hard as you do. Durable, useful, and designed to make a lasting impression. Think quality apparel your team will wear again and again, including popular and exclusive brands. Drinkware that's enjoyed again and again. Bags, notebooks, tools, and tech items that don't just look good, but actually get used.
22:09With thousands of customizable options, 4imprint makes it easy to find what fits your brand and your budget. You'll get expert help, free samples, and their 360-degree guarantee means you can be 4imprint certain your order shows up just right, right on time. Whether you're gearing up for fall events or simply planning ahead for the season, 4imprint can help your brand show up, stay useful, and make connections that last. Explore the possibilities at 4imprint.com. 4imprint. 4certain.
22:43Ed Ludlow:in case you missed it friday spacex has completed its 60 billion acquisition of ai coding startup cursor a key part of elon musk's bid to gain ground on rivals anthropic and open ai the deal became effective on friday according to a regulatory filing two months after spacex formally announced it had agreed to take over the deal is one of the largest ever in the tech industry. Let's stick with SpaceX. The investment firms for ultra-wealthy individuals have built up large exposure to the sought-after AI and rocket company. More than a dozen family offices from across the Americas, Europe, the Middle East, held stakes totaling at least$3.8 billion in the first half of this year, according to filings.
23:22Ed Ludlow:Bloomberg's Devin Pemberton is here with the wealth beat. This is so interesting. You know, the power of the family office and now the positioning of these family offices. Who are we talking about getting into SpaceX here? So we're seeing a huge swath of names. We see Nicholas Pritzker of the Pritzker dynasty. We see Sheikh Mohammed bin Zayed Al Nahyan, who is a royal from the UAE. We see Michael Platt, very well-known UK hedge fund billionaire. Gina Reinhart, who is one of the richest people in Australia. She's a mining magnate. So really just a who's who of billionaires from all over the world.
24:02Ed Ludlow:$3.8 billion is a lot of money, Devin. Is it, though, in the context of like family office positions in other public equities and other big names in the AI space? Well, it really varies. I mean, with Sheikh Mohammed bin Zayan al-Nahin, for example, he was shown to have$65 million, which for his portfolio is quite, quite small. I mean, he's got many, many positions, but that is not that consequential. But for someone like Nick Pritzker, his position was$1.8 billion. And in the context of his overall portfolio, which is probably, you know, around sort of$6 billion at most, I mean, that's really significant.
24:40And that shows he is among those billionaires who are really bullish, not just on SpaceX, but on Musk's entire empire. Pritzker is also an investor in Tesla. and there is a good number of billionaires and family offices out there who are very long on Musk in the whole Musk empire and you can bet that he is a long-term holder.
25:02Ed Ludlow:We'll just recap that SpaceX's IPO is June 11th, first day of trading June 12th, which is an interesting recent history and that$3.8 billion is around a dozen family offices. So there's like concentration there. Would you remind us of methodology? Like how do we get access to this data? Yeah, so these are from 13F filing which are filed at the end of every quarter. And they're showing the positions of money managers that have to report, that have to disclose because they have more than$100 million in U.S. public listed stocks. So, of course, this is just a small number of the family offices out there.
25:39Many, many do not need to file, but they still have vast equity holdings. We just don't have visibility into them. So this is kind of just a slice of the family office universe that we get to see. And it provides lots of interesting data, and we can sort of extrapolate from there what family offices are likely investing in, because we do see an interesting cross-section of family offices from all over the world who've made their money in all sorts of different ways. But it really is just a snapshot.
26:07Ed Ludlow:SpaceX up 5.5 % of the day. Bloomberg's wealth. Devin Pendleton, thank you very much. Take a look at Snap. Okay, Snap is down 3.8%. It was down as much as 15 % in the pre-market and then kind of paired declines. And to be completely honest with you, we don't really know why. And we're trying to find out. There's some squishy trading going on with the social media name this morning. Coming up, Chipmaker, CXMT is leading the way for China's new tech era. We have the details on that halfway through the show. Therefore, it is halftime. Cue beautiful pictures of San Francisco after the markets. This is Bloomberg Tech.
26:59Ed Ludlow:Welcome back to Bloomberg Tech. Look, stocks are trying to find their feet. There is outperformance in technology, in large part because of Bloomberg's report on anthropic second quarter revenue, which we'll give more details on in a sec, but also headlines from the president of the United States about the war in Iran with Brent crude oil hovering around$89 a barrel, all factors in the market to start the week. But there is some big pieces of technology news. Let's get a lot more of that with Bloomberg's cross-asset reporter, Isabel Lee, in New York. Hi. Hi, Ed. Definitely, we have to talk about Anthropic because it's an AI-fueled risk party this morning.
27:33I mean, a lot of the AI chip names are up. Anthropic isn't trading, but it is still the big news. We have a fantastic scoop from our reporters that they told prospective investors that Anthropic saw a 14-fold increase in their revenue compared to the previous quarter. And this is really a big deal because it shows you at a rapid pace of development. We also have Anthropic's annualized revenue, our run rate crossing$47 billion in May. That's compared to OpenAI, which has an annual run rate of over$40 billion. So Anthropic is obviously the clawed chat GPT that all my friends seem to be using. I need to probably use it more, but it just really shows the stunning pace of growth, Ed.
28:12I want to talk next about NVIDIA. Of course, the show wouldn't be complete without NVIDIA. So NVIDIA will invest$1.5 billion in SB Energy and also an exclusive AI compute infrastructure provider at a tech campus in Ohio. So what does this mean? It just means, Ed, that NVIDIA has moved from just selling GPUs and to now financing the factories that produce GPUs. So again, also a really big growth story that you have been following closely in your show. And lastly, I want to talk about Okta. Maybe it's not as talked about, but then Okta is the thing that also backs a lot of these companies. Stock is down around 1.5%.
28:48Even as Wells Fargo upgraded the stock to an outperform. They said that Okta is improving the end market demand and execution could generate a durable low-teens potential growth profile with a limiting downside. So you see the stock there now lower by 1.5%. But the price target of Okta was also raised at RBC and TD Cowan. So really lots of tech news today. And on this Monday morning, back to you.
29:10Ed Ludlow:Thank you, Isabel Lee. Thank you very much. China, a country that once trailed in hardware manufacturing, is experiencing a breakout moment for its chip industry, propelled by the success of CXMT. Now a new tech era is unfolding for the country. Bloomberg's Peter Elstrom. He's our executive editor that leads our coverage of Asian technologies with us. We continue to write about the momentum that CXMT has. But it's a case study, right, of what China wants to do with its domestic chip sector. It is. It is. Ed, we spent so many years talking about the big tech companies in China being Alibaba and Tencent, Baidu, for example.
29:48And now what you're seeing is this chips company, CXMT, which is not that well known outside of the country, went public. And it's already surged beyond all of those companies. So its market cap has gone up to about$613 billion as of today's close. That's$100 billion more than Tencent, the number two there in the market. And as you say, it really signals a changing in the guard. We've had these software companies, services companies like Tencent as the leaders really in the tech industry for many years. And now you're seeing this momentum coming from the hardware side. CXMT, of course, has been one of the companies talked about by Apple in particular.
30:30Apple, with this memory crunch that we've seen out there, wants to be able to buy chips not just from Samsung and SK Hynix and Micron, but also from CXMT. The company has got very strong demand for its memory chips. It's a bit behind those leaders, but it is gaining momentum. It's making a lot of money right now because the prices are so high. And we're seeing further developments in the hardware space, too.
30:52Ed Ludlow:China's focus on its domestic semiconductor industry is multifaceted, right? So we talked about the memory component. There is the chip manufacturing component, trying to get a national champion in the same design as CSMC. you know, the logic side as well. Focus on those two ladder parts and where China is really pushing. Yeah, as you say, it is multifaceted. So CXMT is the domestic leader in memory chips. SMIC is the leader in logic chips at this point. It operates much like a foundry like TSMC. It has made some progress, but because they can't buy the latest machinery from ASML, for example, They're not able to make chips quite as sophisticated as what we've seen out of TSMC in particular.
31:38And it's important to look at some of the other aspects, too. This week, we're going to have the IPO of Unitree Robotics, which is the leader in the robotics space. This is one of many companies in China that are making humanoid robotics. They're coming out there publicly. They're selling quite a bit of them. The Chinese manufacturers have something like 95 percent share in this robotics space. That's another initiative that's a priority for Beijing right now. They want to be able to establish themselves in some of these hardware areas and really build this foundation for the tech industry going forward.
32:09Ed Ludlow:What's the policy story here? How is China supporting that industry? Well, in a lot of different ways. Of course, the government, the economy is very, very different from what we would see in the United States where there's much more government involvement. But they've focused companies on these key areas that they see as strategically important to the country. So manufacturing is very critical. Semiconductors is an area where obviously they're behind what you can get in the West and from Taiwan in particular at this point. But they want to be able to build that foundation. They see opportunities like robotics is another area.
32:44And we haven't even talked about the AI models. I mean, many of those companies are now getting state backing, too, because Beijing sees the strategic importance of AI. And you've got not just DeepSeek and Moonshot, which are both going to go public, by the way, but you've got a whole host of other companies competing there like Jipu and Alibaba. They're competing quite aggressively in this open source space, which is not just good for them. I mean, it helps them offer products at very low prices, but it's also presenting quite a competitive challenge to the likes of Anthropic and OpenAI, despite the revenue figures that we saw earlier.
33:20Ed Ludlow:We're reflecting on a really deeply reported Bloomberg Businessweek story by our colleagues, and I would really recommend you go and read it. Bloomberg Executive Editor Pete Telstrom, thank you very much. Alibaba launched the beta version of Happy Shrimp, an AI music generation model that can turn a single text prompt into a fully produced song. And to boost its AI pivot, Alibaba will sell its gaming arm, Lingxi Games, for$1.5 billion. The firm's undergoing a reorg led by CEO Eddie Wu. making AI and cloud computing a strategic priority to reach$100 billion in AI revenue in the next five years.
33:56Ed Ludlow:Okay, coming up, fully autonomous excavators get to work on construction sites in the US. We're going to speak to the CEO of Bedrock about the rollout next. This is Bloomberg Tech.
34:17The people who seem to get more done than everyone else. They're not working longer hours or running on more caffeine. They've just stopped wasting time on the stuff that doesn't move work forward. Switching apps, re-explaining context, hunting for files. Those aren't small inefficiencies. They're hours wasted every week. Superhuman Go gives you those hours back. From the makers of Grammarly, Go is an AI assistant that sits inside every tab and tool you already use, always available and already aware of what you're working on. Ask it to draft something, summarize a long thread, pull up a file, or prep you for a meeting.
34:58Go handles it without you ever leaving the page you're on. This is what it looks like when AI actually fits into your work, instead of adding to it. It's like having a teammate whose only job is to help you be better at yours. Go keeps up so you can move forward. With Go working with you, you can show off what you do best. Superhuman Go. Find out more at superhuman.com. That's superhuman.com. The thing about AI for business, it may not automatically fit the way your business works. At IBM, we've seen this firsthand. But by embedding AI across HR, IT, and procurement processes, we've reduced costs by millions, slash repetitive tasks, and freed thousands of hours for strategic work.
35:44Now we're helping companies get smarter by putting AI where it actually pays off, deep in the work that moves the business. Let's create smarter business, IBM. It's time to plan ahead and make sure your brand is showing up in ways that can have an impact. That's where 4imprint comes in. 4imprint have promotional products that work as hard as you do. Durable, useful, and designed to make a lasting impression. Think quality apparel your team will wear again and again, including popular and exclusive brands. Drinkware that's enjoyed again and again. Bags, notebooks, tools, and tech items that don't just look good, but actually get used.
36:21With thousands of customizable options, 4imprint makes it easy to find what fits your brand and your budget. You'll get expert help, free samples, and their 360-degree guarantee means you can be 4imprint certain your order shows up just right, right on time. Whether you're gearing up for fall events or simply planning ahead for the season, 4imprint can help your brand show up, stay useful, and make connections that last. Explore the possibilities at 4imprint.com. 4imprint. 4certain.
36:56Ed Ludlow:Let's get over to New York where Bloomberg's Yahaira Anand is standing by. Hi, Yahaira. Hey, Ed. It's time now for Talking Tech with a bit of a defense theme. First up, aerospace manufacturer L3 Harris replaced CEO Christopher Kubasik in a surprise shakeup after finding conduct that was, quote, not consistent with the values of the company. Sam Mehta, a longtime aerospace executive, was appointed CO and member of the board, effective immediately. Plus, Raytheon won a$22.9 billion contract to supply the Navy with Tomahawk cruise missiles. The award comes amid mounting concerns that the war in Iran has depleted the U.S.
37:31military's inventory of precision strike munitions. And Viasat announced its selected space company Rocket Lab to build a high-Earth orbit satellite for the U.S. Space Force. The pair will join forces to deliver maneuverable anti-gham satellites for national security missions. Ed?
37:49Ed Ludlow:Thank you, Yuhaira. Okay, for those who say the future will be built by robots, here's a site for you. The startup Bedrock Robotics has launched the first fully autonomous excavator in the United States at three sites. The debut comes as the AI build-out and other factors are spurring demand for construction. So AI is kind of answering the call here. Boris Soffman, Bedrock Robotics co-founder and CEO, joins us here in SF. Let's get to the basics of what we're talking about. We're talking about excavators, dump trucks, other construction equipment, but there is no driver in the cab. That's correct.
38:23Thank you for having me on, Ed. We're building the Bedrock Operator. This is a suite of sensors, compute, and software that transforms existing heavy machinery, such as excavators and other machines, and enables them to be fully autonomous. And our customers are the contractors who leverage this technology to be able to build faster, safer, and more predictably.
38:45Ed Ludlow:Okay, let's get to the technology. So, you know, analogous with an autonomous vehicle, a robo-taxi, or a self-driving truck, there's a suite of sensors. That's right. Tell me about them. Who offers them, produces them, and then you must have your own software platform too. That's correct. So we use a combination of cameras and LiDAR. It gives us 360-degree visibility around the machine. that's incredibly valuable both for productivity as well as safety, which is one of the biggest challenges in this space. We have compute that is on the machine that's the brain behind operating the system. Who provides the compute?
39:21We use NVIDIA chips for compute, and then we integrate our own systems around it. I'm not surprised by that.
39:28Ed Ludlow:The audience is probably not surprised by that. Yeah, they're still pretty incredible and oftentimes a standard for these sort of systems. Now, what's nice is that that ecosystem has matured a lot, and so we're able to really focus our energies on the software that brings a level of precision and capability that was previously impossible in these sort of spaces. The news here is that this is real. You've deployed in three commercial sites, right? How real? I mean, are you able to book revenues on these three sites? In other words, someone's actually paying you for this. And what's the scale of the deployment?
40:00Yeah, we are. So we are on three different sites with three different partners. Sun Construction, Zachary, Champion Site Prep. There are three different projects, everything from water treatment facilities to other types of infrastructure. These are paid deployments that are doing real work that's on the critical path of these projects. And these are projects that are doing up to multiple millions of cubic yards of earthwork as a project, and we're doing a subset of that.
40:30Ed Ludlow:So now is the big question. Why is there a need for autonomous construction equipment? It's a great question. There is an unprecedented need to build. So we're unshoring manufacturing, there's housing shortages, road shortages, the data center boom that we've all seen, water treatment facilities. All of these are pushing on an already incredibly strained industry. And the labor supply is actually continuing to fall where some of the contractors we partner with are seeing huge retirement rates. And so there's an absolute bottleneck that effectively becomes a radiation on the whole economy when you can't build the things that the country needs.
41:08Ed Ludlow:So in very simple terms, there are not enough drivers nor operators to be inside the cab of any piece of construction equipment. That's right. And some of these are actually incredibly complex pieces of equipment, like an excavator operator needs many years to actually become competent because these are incredibly complex, sophisticated machines. What about the utilization rate then? What can you achieve in a 12-hour shift with an autonomous piece of equipment relative to a human-driven? Yeah, well, you nailed it. So we're already close to human productivity. We're going to reach human productivity very soon.
41:40Ed Ludlow:What is the benchmark of human productivity in a construction site? So, for example, the volume of Earth moved in a period of time. Right. But you nailed a really important point that already these machines are oftentimes sitting idle. Contractors are turning down work. because they physically can't reach the demand with the capacity that they have. And so this allows you not only to fulfill that, but you can actually now start compressing schedules and doing more work in a physical block of time by exceeding what typically would be one shift and working longer stretches that actually can compress schedules.
42:15Ed Ludlow:Team, let's bring the pictures back up, because I think you can't get enough of showing it in action. That's quite a sight. That is a well-known brand of construction equipment. I'm sure others are available. But it reminds me of a period of time where I was covering the EV industry. There was a section of that industry that just wanted to retrofit the powertrain of existing commercial vehicles. So you must have to partner with those OEMs and add your technology stack on top. Well, the interesting thing is that a lot of these machines are actually designed incredibly well to be able to be transformed into autonomously capable and to do it safely without a major redesign.
42:55That's not the case in public road cars and trucks. There's specific aspects of these machines that actually make them much more suitable for this sort of automation.
43:04Ed Ludlow:So what is the constraint on your ability to build the fleet that you can then deploy? Because the number of projects you take on will be dictated by how many of these autonomous machines can be deployed. That's exactly right. And so this is where it helps to have millions of these machines already out there that become part of the pool that you can go and pursue. And just like we saw in other big waves of technology like this, such as autonomous driving, There's a really deep challenge in the zero to one where you're trying to deploy a system safely and productively. And then there's another suite of challenges when you start to scale it into the hundreds and thousands and beyond.
43:41And so now it's a matter of really maturing the autonomy capabilities and starting this flywheel to move into new capabilities and new machines and then be able to really orchestrate them where you can start to complete more cohesive sets of work.
43:54Ed Ludlow:But in simple terms, you take delivery of the piece of equipment, you retrofit it with your stack, and then you put it out in the real world. Even better, we do it on site. It takes us a few hours. It's 100 % reversible, and it's ready to go as soon as it's complete. So it's a really incredible kind of ability to start to really deploy and scale this. So the way that we describe the flywheel at the top of the show is, you know, there is AI-driven demand for the machinery, and AI is answering the call. Talk about some of the data center projects. You know, the hyperscalers themselves that end up renting or owning the capacity directly looking to you, or is there still a degree of separation with the contractor?
44:33Ed Ludlow:You know, they must have an interest in this field. Yeah, our direct customers are the contractors, and we pursue the work that is most valuable to them. That happens to obviously have a big pool in data centers, but we also see projects of all sorts, highways, water treatment facilities, housing developments. And so we end up enabling them to take on more of this work and be able to scale with those demands. And oftentimes we do see this push actually not from our direct customers, the contractors, but the end owners that actually get a huge benefit in the predictability. Is there any end owner that's particularly interested in you at the moment?
45:10There's a large number from the broader ecosystem and space. And it's actually both end customers as well as developers, whereas you can imagine one of the really interesting impacts of this is that if you can compress schedules, the types of projects that previously didn't pencil out because of the price of loans and so forth suddenly do. And so this actually becomes incredibly expansionary for commercial developments and housing developments that previously would be out of reach.
45:35Ed Ludlow:Very interested in your commercial progress. In autonomous haulage or trucking, you can do a very easy comparison, dollar per mile, against a human-driven truck. Where are you at dollar per ton of waste moved, or is it done by time against the human-operated machine? We've explored a few options with our partners, and we're figuring this out actively. Sometimes this is a license where it becomes like an operator that you've brought on to join your fleet. Some have done it by productivity, like volumes of cubic yards that are moved. What's nice is that one of the biggest challenges that contractors face is uncertainty of the length of projects.
46:14This actually allows them to tighten that up, which gives them huge flexibility in taking on more work and tightening up the windows in which they do it.
46:20Ed Ludlow:Autonomous construction equipment. Bedrock CEO Boris Hoffman. Thank you very much indeed. Okay, coming up, two groups with different ideas on how to regulate AI are backing the same candidate for Florida Governor. We'll have the details next. This is Bloomberg Tech.
46:49Ed Ludlow:The CFTC is taking a closer look at one of AI's newest markets, tradable computing power. The market's watchdog wants to take public comment on contracts tied to compute capacity, according to a posting made public Monday. The move signals there are still lingering questions about the new tradable asset class, and it could delay plans for a futures market and one of AI's most critical resources. A battle to influence AI regulation is playing out in Florida in an unusual way. Groups with rival visions for AI oversight are both backing Republican gubernatorial candidate Byron Donalds. He leads in the polls for the primary tomorrow and the general election coming in November.
47:31Ed Ludlow:Bloomberg's Emily Birnbaum joins us now. OK, really, really simple. There are two groups with each their own vision of how to regulate AI. They're both backing the same candidate. How does the candidate feel about it? So, the candidate really interestingly has struck an ambiguous tone. So, the big difference between these two AI super PACs is one believes each state should be able to regulate AI as intensively as they are able to, and the other believes there should be one national light touch standard. And somehow Byron Donald, I guess a pretty good politician, has been able to speak in a way that makes both sides feel he's on their side.
48:16So essentially he has said he believes that the U.S. Congress should lead the way on passing AI regulation. But because the U.S. Congress is so dysfunctional that states do have some sort of role to play, especially when it comes to protecting children from potential AI harms.
48:33Ed Ludlow:You know, the logic here is that were he to to win, you know, he would be influential, right, and able to to have an impact at the whatever layer in getting these things move forward. Yes. So he is going to be if he wins, he would be filling the role of current Florida Governor Ron DeSantis, who's been a leading conservative voice in favor of AI regulation. So DeSantis tried to push forward what he called an AI Bill of Rights. It failed in the Florida legislature. But Byron Donald is going to choose whether he's going to continue with that vision and continue to be a pro-AI regulation Republican or whether he's going to strike a more business friendly tone.
49:15Ed Ludlow:Team, let's show that table again of the backers of each organization. The umbrella point is that both organizations have spent some serious money in support of Byron Donalds. On the favoring light regulation side, Mark Andreessen of Andreessen Horowitz, Greg Brockman of OpenAI. On the AVI safety, Dario Amadei, CEO of Anthropic. How serious are these campaigns? Real quick. They are very, very serious. They are spending tens of millions of dollars across the country, going head to head in some instances, playing in different states and others. They both see this midterm cycle as key to the future of AI regulation, and they are not afraid to spend.
50:00Ed Ludlow:Bloomberg's Emily Birnbaum, terrific reporting on what's happening in that Florida race. Thank you very much indeed. That does it for this edition of Bloomberg Tech. Another look at markets again, technology outperforming. The president's comments this morning on Iran and Oman, both factors, oil hovering around$89 a barrel on Brent. And then there's the impact of that anthropic revenue number for the second quarter, which is driving upside in some AI stocks. What a way to start the week. Recap on the podcast. You know where to find it on the Bloomberg Terminal as well as online on Apple, Spotify and iHeart.
50:34Ed Ludlow:Have a great week. Stay with us throughout it. This is Bloomberg Tech.
50:41When 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. At 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 the Hartford.com slash risk mitigation.
51:17Policies provided by Hartford Fire Insurance Company and its property and casualty affiliates, Hartford, Connecticut. As industries evolve faster than ever, companies need an environment that accelerates strategic growth. And Michigan delivers on that promise. From emerging startups to global enterprises, Michigan offers what executives value most, a resilient, innovative ecosystem, diverse communities that attract top talent, and a quality of life that supports work-life balance. With our unified Team Michigan approach, businesses scale faster and compete at the highest level. Michigan, pure opportunity.
51:50Seize your opportunity at michiganbusiness.org. It's time to plan ahead and make sure your brand is showing up in ways that can have an impact. 4imprint's promotional products are designed to work as hard as you do and make a lasting impression. From quality apparel, including exclusive brands, to drinkware, tech, and totes, they've got thousands of options to fit your brand and budget. Plus, you get free samples, expert help, and their 360-degree guarantee. So you can be 4imprint certain everything shows up just right, right on time. Explore more at 4imprint.com. 4imprint. 4certain.
From the publisher
Bloomberg’s Ed Ludlow breaks down Anthropic's second-quarter revenue jump to more than $11.5 Billion, 14 times more compared to the same period a year ago. Plus, some family offices from around the world have built up billions of dollars of exposure to SpaceX; and Bedrock Robotics launched the first fully autonomous excavator in the US.
See omnystudio.com/listener for privacy information.

