

22 Oct 2026 · 46m · 18 chapters

Bloomberg Tech covers Meta’s AI spending and earnings, including a new non-voting Class C share structure for Mark Zuckerberg; Tesla’s profit beat alongside reduced CapEx and implications for FSD/RoboTaxi; IBM’s mixed results; and an AI-driven cybersecurity case study involving a major PHP vulnerability.
Greg Brockman (OpenAI President) and Matthew Prince (Cloudflare CEO); plus analysts Alex Levine (Bloomberg), Minda Smiley (eMarketer), Ben Harwood (New Street Research), Fatima Bulani (Citi), Alison Porter (Janice Henderson).
Meta beat Q3 and guided strongly for Q4, narrowed CapEx to $135B–$145B, and says AI agents (Muse/Muse Wonder) show early traction (10M signups in 6 weeks; 500M+ businesses using its apps). The Class C shares are framed as funding/flexibility for AI spend and tied to Zuckerberg staying in California/leadership for eight years while preserving voting control. Tesla: FSD adoption at 75% of US vehicle sales; CapEx cut raises questions about AI/autonomy pace. Cybersecurity: OpenAI models helped find a more complete PHP exploit; Cloudflare reports millions of attempted attacks per minute and urges updating PHP.
Muse monetization via shopping use case (~quarter of users); Tesla potential SpaceX-Tesla merger synergies; IBM “AI survivor” framing; PHP flaw protection and responsible disclosure.
Written by AI. May contain mistakes.
Discussion on Meta's creation of non-voting shares and leadership stability.
“Shares down almost 7 % in the session on track for their biggest drop since July.”
Examination of Meta's strong earnings and AI-related product launches.
“But she didn't mention the Zuckerberg part of it at all.”
Insights into the adoption rates of Muse and potential revenue strategies.
“We are building out the capacity to serve more than a billion people over the next few years.”
How AI is enhancing Meta's advertising capabilities and revenue growth.
“It was still the core ads business that drove the beat, right?”
Discussion on Zuckerberg's commitment to Meta's AI future and leadership.
“Minda, I'm not going to ask you to come in here on the mechanics of them issuing a new class of stock.”
Analysis of Tesla's earnings beat and the implications of spending cuts.
“They were down more than 3 % a little earlier.”
Discussion on the potential merger between Tesla and SpaceX, including its implications and market dynamics.
“But everything that we saw on the quarter was backing up our price target.”
Exploration of the regulatory environment and potential challenges for the Tesla-SpaceX merger.
“I want to say on that potential merger, Bloomberg's learned there is a push to announce a deal by the first half of next year, with Musk and his advisors looking to get it done while President Trump is still in office.”
Analysis of IBM's earnings report, discussing revenue growth and investor concerns amidst the AI landscape.
“IBM delivered better than expected sales growth in its third quarter earnings report, but consulting revenues missed estimates.”
Discussion on Meta's new class C shares and the implications for the company's market position and credit risk.
“AI risk, I would say, is kind of the prevailing theme of the market this morning.”
Learn about Meta's introduction of a new class of stock and its implications.
“I'd say also you have NVIDIA, AMD, et cetera, as the big drags at the index level.”
Discover how Meta's CapEx commitments relate to their AI spending strategy.
“And I should say that in the earnings statement and on the call, meta CFO Susan Lee framed this very much as a mechanism to get flexibility in financing the AI spend.”
Explore the evolving market strategies around AI investments and infrastructure.
“But the direction of travel is pretty clear.”
Understand the circulation of free cash flow within the tech ecosystem and its implications.
“to be very active because this volatility is here to stay.”
Learn how AI is both helping to discover and potentially exploit security vulnerabilities in PHP.
“It's a programming language used to build and run millions of websites and web applications.”
Explore the responsible actions taken by OpenAI and Cloudflare after discovering vulnerabilities.
“They disclosed it to us in a confidential way.”
Understand the concept of the 'defender's window' and its implications for internet security.
“and we're going through code and we're going to find a bunch of flaws.”
Discover how AI tools like Dots are improving productivity and protecting users from threats.
“Greg, before I let you go, I've been using Dots a lot.”

Get the gist, explore an idea or ask about something said.
On Bloomberg Tech, what are the key points of this episode?
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. Some people treat ChatGPT like some kind of smart search engine, and some use it to get work done. ChatGPT Work is a new way of working in ChatGPT that can take action across your apps and files, stay with a project for hours if needed, and turn a goal into finished work.
0:46It's designed to help you move from a chaotic starting point to a reviewable first version. So all the source materials, briefs, and scattered information that you have to grind through to turn into something useful can just become something useful. Put ChatGPT to work on your most ambitious ideas and projects. Get started at ChatGPT.com by selecting Work Mode, available on Plus and Pro plans. Small businesses are the pulse of every community. They bring people together, create opportunities, and drive growth. Chase for Business helps business owners like you with personalized guidance and convenient digital tools all in one place.
1:24With that guidance and your determination, you can take your business farther and help build a brighter future for your community. Learn more at chase.com slash business. Chase for Business. Make more of what's yours. The Chase Mobile app is available for select mobile devices. Message and data rates may apply. JPMorgan Chase Bank N.A. Member FDIC. Copyright 2026, JPMorgan Chase and Company.
1:50Bloomberg Audio Studios. Podcasts, radio, news.
2:01Bloomberg Tech is live from the heart of Silicon Valley with Ed Ludlow in San Francisco.
2:10Ed Ludlow: This is Bloomberg Tech. Coming up, meta doubles down on AI, boosting its spending outlook, as Zuckerberg says its massive investments are starting to pay off. Plus, Tesla beats on profit, but a big cut to its spending plans raises new questions about Musk's AI, RoboTaxi, and robotics ambitions. And AI reshapes cybersecurity for both defenders and hackers. We speak with OpenAI President Greg Brockman and Cloudflare CEO Matthew Prince. Let's get right to our top story, and that is Meta. Shares down almost 7 % in the session on track for their biggest drop since July. I'll show you the two-day chart because going into earnings, there was a lot of buoyancy.
2:46Ed Ludlow: It's dragging down technology more broadly. A lot of the chip names also a big drag at the index level. Meta is creating a new class of non-voting shares. And part of the reason is Mark Zuckerberg. Meta says the CEO was considering leaving California over a proposed wealth tax that could leave him with a multi-billion dollar Bill, the new Class C shares gives Zuckerberg stock he can sell for cash without immediately reducing his voting control. In return, he's agreed to stay in California or wherever Meta is headquartered for eight years, as long as he remains in a full-time leadership role. That was one story.
3:21Ed Ludlow: Let's dive deeper into what else we learned from Meta's earnings report, which was yesterday. Bloomberg's Alex Levine is with us. Let's be completely frank about it. The Class C shares story was communicated in a difficult to understand way, partly on the call, partly in the earnings statement, but really in a filing. And then there were lots of numbers from earnings. We'll start there. What do we actually learn about how it's going for Meta right now? Well, we know that Meta delivered really strong third quarter results, really strong fourth quarter guidance. They beat analyst estimates on those.
3:51And I think that the decision to issue this non-voting stock and Zuckerberg's decision to stay in Silicon Valley to sort of see everything through is a really important statement as Meta enters this next chapter, continues to accelerate into this next chapter. And I think it also just goes to signal that it is confident that its AI bets are paying off because some of the decision to do this with the shares was, of course, also, as the company had said, to help cover some of this enormous AI spending down the line.
4:23Ed Ludlow: Yes, Meta's CFO Susan Lee was asked on the analyst call about the class she says, and she said it is a way of diversifying and funding and financing the AI spend that's to come. But she didn't mention the Zuckerberg part of it at all. CapEx is narrowed for this year, 135 billion at the bottom end to 145 billion at the top end, which we knew. There seems to be like some signals on the direction of travel for Meta. AI seems to be also really boosting the core ad business. What did they say? What were the numbers that they'd point to to evidence that? Well, just in the last month, Meta launched these two new products.
5:04You had its personal AI agent, Muse, that was debuted in September. And then you had Muse Wonder, its latest AI model that debuted earlier this month. And what Meta reported, it was really strong early adoption, strong early traction in all of this. And I think that even even though investors do still seem skeptical, as you said, shares fell this morning, even as Zuckerberg and the company projected all this confidence. I think that they are really pointing to the early traction of these products and also the huge number of businesses that are already using their other, you know, the other platforms across their family of apps.
5:38Instagram, WhatsApp, Facebook, I think the number that they said was more than 500 million businesses. They're really pointing to these numbers as evidence that news adoption will continue to scale and that it can actually be monetized, which remains the big outstanding question here is how are they actually going to really make money off of all of this?
5:55Ed Ludlow: The shares are down six and a half percent on track for the biggest drop since July. But what was weird is when all of this news came out last night, the stock really went up and things really changed overnight. Bloomberg's Alexandra Levine, thank you so much for being with us. During Meta's earnings call, Mark Zuckerberg touted the progress of its AI agents, or Muse. We are building out the capacity to serve more than a billion people over the next few years. In the near term, Muse is very compute intensive to run, but we expect to improve its cost efficiency by more than 10x in the next year, which will enable it to be profitable once the revenue lines are at scale.
6:33Ed Ludlow: Let's get more meta's results with Minda Smiley, senior analyst at eMarketer. And I want to start with Muse. What they also said is that more than 10 million people signed up for Muse in the US and Canada in its first six weeks. How do you interpret that data? Yeah, I mean, that's certainly a good number in terms of early adoption. But I do think when you think about the fact that a lot of these downloads and usage are coming from the fact that Meta has been able to promote Muse through its own channels, Facebook, WhatsApp, Instagram, quite heavily. And they've also just invested a lot of money into marketing Muse as well.
7:07So I think it kind of waters down those numbers in a way because I think it points to the fact that a lot of this interest isn't necessarily organic, isn't necessarily genuine. It's really just coming from the fact that people maybe might see something about Muse and Instagram, click on it, download it, mess around with it. And then maybe a week later, they delete it from their phone.
7:27Ed Ludlow: I've been using Muse as much as I can, like as somebody interested in technology. If you host a technology news show, you should probably use the technology right there. I'm still kind of not really using it beyond helping put social posts together. More deep search, maybe it's helping me with my calendar. So I found it interesting that Mark Zuckerberg said that they expect most people to use the free version. The monetization comes elsewhere. Is there any evidence that they've got going on the elsewhere? Yeah, well, it's interesting. I mean, they talk a lot about how they hope to maybe collect fees from Muse as they help with transactions.
8:01One of the stats they pointed out yesterday was that about about a quarter of users or I think they said the main use case at this moment is shopping, which, of course, you know, as a company that's invested quite a lot of money into social commerce, it makes sense that they would want Muse to also be a commerce tool. So I think when it comes to monetization, there are a lot of opportunities there. When you think about the fees, when you think about subscriptions, even extending Muse to small businesses. But I think a lot of that still remains quite theoretical, right? I mean, there's no guarantee that this will take off.
8:36It has to really gain traction with consumers, with businesses before those monetization lines can really go into effect.
8:43Ed Ludlow: It was still the core ads business that drove the beat, right? I get the stock is down significantly, but ad revenues were more than$63 billion. The street expected below 62. And the interesting point is price per ad growing at a rate of about 14 percent was also beyond expectations. It seems like the AI story really is that they're able to drive higher pricing on ads rather than look at volumes. Yeah, it's funny because I think because there's been so much focus on, well, where are they going with AI? What are their ambitions? Everyone wants to hear about Muse. And it makes sense, right? They're spending a lot of money.
9:18We want to see where they're going with all of this. But the reality is that AI is really fueling their advertising business, which we all know is the core, the bulk of their business model. And it's doing so quite effectively. That's why we continue to see this double digit growth quarter after quarter. And they keep refining it, keep building on it. And it's helping not only with advertising, but also usage and engagement as well.
9:41Ed Ludlow: Minda, I'm not going to ask you to come in here on the mechanics of them issuing a new class of stock. The whole point is that it's structured so that Mark Zuckerberg can sell 20 % straight away and cover his tax bill if that were to arise. And 80 % is staggered over eight years. But that's the bit I want to talk to you about. Mark Zuckerberg has agreed to stay and enroll basically for eight years, wherever the company's headquartered, California or otherwise. How important is that? You know, the founder CEO continuing to lead efforts like AI. Yeah, it is important. And it's a question I've been getting recently.
10:15You know, what does this all say about Zuckerberg and his commitment to the company? If, when will he step down? And I think it just goes to show that he is committed to this vision. And, you know, to what extent it will play out, of course, is a different story. Obviously, we see that we saw what happened with the metaverse. We see that he's going heavily into smart glasses as well, which doesn't get as hasn't gotten as much coverage lately with the launch of Muse. But I think it goes to show that he doesn't plan on going anywhere anytime soon.
10:46Ed Ludlow: Minda Smiley, senior analyst at eMarketer. It's been great to have you back on the program. Thank you very much. Coming up, the other earnings story, which was Tesla beats on profit, but cuts billions from its spending plans. What that means for Musk's AI, robo-taxi, and robotics ambitions. We're going to look at Tesla next. This is Bloomberg Tech.
11:08Ed Ludlow: Okay, shares of Tesla down, but way off session lows. They were down more than 3 % a little earlier. The company beat on profit, but it's the spending outlook that's getting investors' attention. It is cutting its CapEx plan by more than$5 billion, raising questions about the pace of its AI, robo-taxi, and optimist ambitions. Joining us now is Ben Harwood, New Street Research Technology and Space Infrastructure Senior Analyst. We'll get to the SpaceX Tesla merger idea of it all. I mean, yeah, it's really interesting. CapEx is being trimmed. Maybe that's just like the reality of how quick they didn't deploy.
11:43Ed Ludlow: Does it signal something bigger to you? No. Hi, thanks for having me on. And the client CapEx I see is just mainly a timing issue. And I expect them to ramp up CapEx, you know, coming into 27 and into 2028. And you mentioned what they're doing on the AI front with FSD, with RoboTaxi. Actually, everything we saw from that perspective was very positive. So what we saw is FSD adoption reached again an all-time high. 75 % of vehicles in the US are now being sold with FSD. And that's really showing that the new technology that you have with FSD is driving the modern. And RoboTaxi is the fastest scaling RoboTaxi business out there.
12:24Ed Ludlow: Ben, let me jump in. because the audience knows I use FSD every single day for my 60 mile round trip commute. I use it, I pay for it myself. But a new Tesla, and I know a lot of the numbers that quarter were existing Tesla owners upgrading. But if you give Tesla somebody FSD for free, they might use it for a while. The question is, do they stick with it? Right? The take rate, the actual long term usage. Do you have a good answer on that? Yeah, exactly. So that's a good question. So every Tesla owner in the US gets a month for free. What the CFO clarified on the call yesterday when we asked the question is actually 75 % of people are buying the technology even when they get it for free.
13:08Ed Ludlow: So, they'll take it in that first month. So, it's very clear that there is a lot of demand for this product. Word is getting out. Like I said, 75 % of new vehicles are being sold with FSD. That's up 20 points sequentially, and that's going to be up 40 points year on year. So, performance of FSD, as you would have noted, Ed, has improved over the past 12 months, and that's really spreading the word, and it's now driving vehicle demand in the US. It's only a US phenomenon now. Hopefully, in the first half of next year, that technology will be approved in Europe, and then we'll start to see similar numbers coming out of Europe and a strong growth in vehicle demand there.
13:49Ed Ludlow: Ben, very quickly, did you change your price target or call on the stock after last night? No. So we're still seeing at 600. So it's a very healthy upside from where we are today. But everything that we saw on the quarter was backing up our price target. It's all based upon Tesla becoming an autonomy leader and exactly what was announced yesterday. Okay, let's talk about special committees at both Tesla and SpaceX looking at the idea of a merger. Last night, Tesla IR said they will not be taking questions on the subject. But 24 hours before that, Musk disclosed it after he was asked on SpaceX's call.
14:28Ed Ludlow: You cover both. You cover the area. What are you modeling for? That transaction, how it looks like, who acquires who, and the prices involved? Yep, exactly. So we think the acquisition is going to go ahead. It will likely happen by the end of next year and it will be announced by the end of next year so that both the companies and Musk himself can take advantage of the political backdrop that we have right now. I think the rose to reveal last week really goes to show the fact that there's a lot of collaboration going on between the businesses. I think you'll have SpaceX acquire Tesla given the fact that Musk has a much higher voting control in SpaceX.
15:09Ed Ludlow: versus Tesla. The Tesla shareholders will want a premium for the acquisition. But there are clear synergies. The Roadster event, as I said, was one of them. But Tesla is a power company. What does SpaceX want? SpaceX wants to roll out data centers and it's powerful that it will leverage Tesla. That's the bit that I'm trying to understand. I'm sorry to jump in. Tesla's data center near-term ambitions 400 megawatts the reach goal for spacex next year is 20 gigawatts like those are different scales of commitment um you know but he talks about the shared goals that they have it seems like one supporting the other yeah it's it's i would say it's it's it's almost two ways so so tesla can uh leverage the compute of spacex spacex will give uh tesla compute without uh enable TESA to improve FSD even faster and ramp up the robo-taxi fleet and performance even faster.
16:05Ed Ludlow: TESA will give the Megapack expertise, solar expertise, so SpaceX can roll out its data centers. Even behind the scenes on the rocket front, TESA provides power components to ensure that they can solve problems on the rocket side of things. But of course, a big objective is for Musk to control one entity as well. But we do see synergies playing out between the two businesses. Ben Harwood of New Street Research. Great to have you on Bloomberg Tech. Thank you very much. I want to say on that potential merger, Bloomberg's learned there is a push to announce a deal by the first half of next year, with Musk and his advisors looking to get it done while President Trump is still in office.
16:49Ed Ludlow: That's all according to sources. And it's the big focus of today's big take. Bloomberg's Craig Trudell, who leads our coverage of both companies, is with us. That's the idea, right? It's the environment, as Ben just outlined, that the companies find themselves in. And there is kind of a deadline to get something done. Yeah, that's absolutely right, Ed. And I think if we take a step back here, these are two boards stacked with a lot of directors who are quite close with Elon, who go back a long time. Tesla's board has managed to pull off, lining up a pretty incredible pay package for its CEO only just a year ago.
17:26So if and when Elon says jump, there's been a sort of track record there of them asking how high. So is that an issue? Probably not. The question I think here is, will there be an administration to sort of review the prospect of a merger involving such big companies that is going to be supportive and I think Musk understandably wants to try and make that push while Trump is still in office.
17:51Ed Ludlow: Well, even nearer term, right, we're a couple of weeks away from the midterms, and so from sort of congressional oversight, the makeup of Congress could change. Like, we talk about that in the big take. Yeah, that's exactly right, and it's not necessarily the case that Congress has all that much of a say here, but it absolutely is the case that, you know, if you're Musk, you're also bracing for the possibility that this could take quite some time. The U.S., of course, will have a say, but it's not the only country that will have a say. And the fact that China looms so large for Tesla with such a presence there, and there's some potential conflict there, right, where the fact that SpaceX is a U.S.
18:30contractor, a U.S. defense contractor, Starlink doesn't necessarily align with how the CCP kind of wants their internet to work. Will they have notes and have a view as to whether or not this merger should go through?
18:42Ed Ludlow: I think that's one of the big questions here. I would point out in a regulatory filing, they confirmed this morning, the special committees, I know it was discussed on the call, and they had the same language, like this isn't a certainty. Just the other thing, of course, is the voting power differences. We're going to get to that in the coming weeks. Bloomberg's Craig Trudell, really great to have you with us. Thank you. Now, coming up, IBM, another earnings story. Delivered better than expected third quarter results overall, but there's still some concern out there. The stock a little weaker, down 7 tenths of 1%.
19:09Ed Ludlow: We've got more on those earnings when we return. This is Bloomberg Tech.
19:19The 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.
20:00Go 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. Some people treat ChatGPT like some kind of smart search engine, and some use it to get work done. ChatGPT Work is a new way of working in ChatGPT that can take action across your apps and files, stay with a project for hours if needed, and turn a goal into finished work.
20:43It's designed to help you move from a chaotic starting point to a reviewable first version. So all the source materials, briefs, and scattered information that you have to grind through to turn into something useful can just become something useful. Put ChatGPT to work on your most ambitious ideas and projects. Get started at ChatGPT.com by selecting Work Mode, available on Plus and Pro plans. The world of business is constantly evolving. Every day, new challenges, new opportunities, new ways of working. Comcast Business keeps you totally in step with secure AI-backed networking built to power the way modern business gets things done today in more than 100 countries around the world.
21:27That's some serious muscle. And get this, Comcast business powers over 90 % of the Fortune 500 and millions of small businesses. Behind it all, network engineers, cybersecurity specialists, support teams, thousands of experts answering your call at 2 a.m. like it's 2 p.m. Always on, always ready. That's a lot of coffee. One partner powering how business gets done for companies of every size all around the globe. When you add it all up, no one does business like Comcast business.
22:26Download it at whisperflow.ai. That's W-I-S-P-R flow dot A-I.
22:38Ed Ludlow: IBM delivered better than expected sales growth in its third quarter earnings report, but consulting revenues missed estimates. Ultimately, the numbers have failed to impress investors. There's some concern out there about this shift in corporate technology spending. Join us to talk more about the earnings. It's Fatima Bulani, managing director at Citi. There's a buy rating on the stock with a target price of$280. It's interesting because a decline of 7 tenths of a percent is not a lot to write home about. And a lot of what IBM presented was quite good. There seems to be some worry out there in the market that IBM is just not it right now in the world of technology spending.
23:14Yeah, let me unpack a lot of that because I think what you're alluding to is IBM is actually a very polarizing stock. Now, just the forensic analysis on the three key print last night, really the headline beats coming from the infrastructure and software segments of the overall businesses, which in totality are two thirds of the overall revenue mix. Those are important points to highlight because we saw a reacceleration in those segments. And they indicate to us one very important thing, that the mainframe business is at the very least stable. Now, you know, in a lot of ways, we thought this was a little bit of a surprising redemption quarter on the back of what was a pretty monumental miss for this business back in 2Q.
24:02So the fact that this company was able to recapture and recoup two thirds of these slipped transactions within a matter of 90 days, we do think that's impressive. But, you know, in the same breath, this is not sort of a decisive victory for the bulls where there is still going to be a lot of work and numbers and execution that IBM is going to have to put up to sort of compel the bears to back off from this narrative around, hey, the mainframe business is under structural duress and existential duress from the AI onslaught from the labs.
24:39Ed Ludlow: That's my question. Like, what is IBM in the AI age? Bloomberg's Romain Bostic got on the phone with IBM CEO Arvind Krishna. And, you know, he presents that, you know, in the shift to inference, IBM is primed for the AI age. Is that a part of your model? Yeah, look, there's a couple of angles to this, right? Our framing here has been IBM is not quite in the AI loser camp, nor are they decisively in the AI winner camp. They are an AI survivor. And we have to remember that IBM is so pervasive in some of the largest and most complex, most heterogeneous IT organizations in the world. So it gives them a pretty important seat at the table in terms of enabling.
25:24Ed Ludlow: You said AI survivor. AI survivor. It's not a bad place to be. Not a bad place to be. I mean, I'm looking at you and your industry colleagues and what in aggregate you're saying. You know, Red Hat drove quite a lot of growth on the software side in the quarter gone. I think a lot of people are basically saying how sustainable is the growth that they just printed going further afield. What's your view on that? So the view here is just to kind of close the thought on this AI survivor notion. It's they are very deeply embedded in very important and consequential parts of the IT stack. Right. So while they're not necessarily getting the torque that you would see from, you know, other AI winners that have been anointed, you know, we think of Datadog in our coverage universe.
26:09We think of a Cloudflare in our coverage universe and even some of the cyber names. Right. So there isn't a clear and unequivocal tether to how they're benefiting from AI. But certainly there's opportunities here from the expanse of IBM's portfolio to help the average organization operationalize and productionize AI. And, you know, that's certainly been the message from the company. And where some of that does fall short in terms of the reality is the organic software business. Right. That is definitely been, you know, a flashpoint for a lot of investors. Right. That's part of the reason why I've mentioned that this is still a pretty polarizing stock.
26:47in spite of the fact that they put up a pretty strong beat, re-accelerated sequentially, and sort of proved to you that they took what was theirs in terms of the mainframe business, right? So, a little bit of a redemption quarter, but not out of the woods yet as it relates to some of the bearish concerns that, hey, maybe they've won the skirmish, but not the war.
27:10Ed Ludlow: The stock had been down almost 5%. You know, when the market opened up this Thursday morning, it's now off by a percentage point as people digest everything that Fatima Bulani, Managing Director of City, just went through. Great to have you on the show. Thank you so much. Coming up in the program, Alison Porter of Janice Henderson joins us. Meta is a top holding of the firm's tech leaders, right? And we'll get her take, yes, on the AI spending. But I want to talk about this new Class C share issuance and the Zuckerberg of it all. Halftime here in San Francisco. Stay with us. This is Bloomberg Tech.
27:45Ed Ludlow: AI risk, I would say, is kind of the prevailing theme of the market this morning. The Nasdaq 100 is off by 2.4 % at its session low, was on track for its biggest drop, I think, since July. The semiconductor index, or the SOX, also is kind of where the brunt of selling is right now. I'd say also you have NVIDIA, AMD, et cetera, as the big drags at the index level. But the earning story, the big one is Meta. Down, again, on track for its biggest drop since July, off 6 % or so, off-session lows. Narrowed the CapEx range, strong top-line growth, the ad business outperforming, good signals for 27, but they issued a new class of stock.
28:24Ed Ludlow: There's also a warning in the credit market. I'm looking at Meta 5-year CDS, which, as you can see on the right-hand side of that screen, is at a record level since trading began. And the credit risk for Meta is becoming a story. We can get into the specifics. Alison Porter, Portfolio Manager at Janice Henderson on the Global Technology Leaders team. I looked at some of your holdings. Meta is up there, right? I would love to start with this new class of stock. It has economic value, but no voting power. And the reality of it is that the company is doing this to give Mr. Zuckerberg some liquidity to cover a tax bill if here in California Prop 40 passes.
29:03Ed Ludlow: but it also keeps him focused and imposed. What do you make of all that? Yeah, I think the strength that Meta has seen on revenue needs to be funded from CapEx and the market has been expecting that we would see a capital raise at some point. And this new share class structure is designed to allow them to focus on potential acquisitions, partly for hiring for employee comp. We saw CJ Desai hired from MongoDB and also to fund CapEx. So it's multipurpose. And we do think this is going to be an overhang on the stock through until we see that capital raise actually occur. So I should say... Potentially sometime in Q1.
29:54Ed Ludlow: Right. It's a 90-day process. And I should say that in the earnings statement and on the call, meta CFO Susan Lee framed this very much as a mechanism to get flexibility in financing the AI spend. But in the regulatory filing, it's also detailed that there's an ad hoc founder led agreement with Mr. Zuckerberg. Is it important to you that you have the founder CEO sort of locked in to lead an increasingly AI focused strategy at the company? Yeah, I think it's very important that we do see Mark Zuckerberg and we do see him continue to lead, like all founder led businesses, including Tesla. That's where we've seen founders be able to have that vision and invest against future growth.
30:41obviously Mark Zuckerberg's being core to that story and having the vision on news driving towards the personal intelligence product but we do think it's really important that they focus on shareholder returns and on the call last night that was a really big feature focusing on you know spending on capex where you actually would see a return and a lot of focus on granular monetization as part of that. But yeah, it is important that we see Mark continue to lead Meta forward.
31:16Ed Ludlow: You mean monetization in the sense of monetizing the AI tools and AI products specifically, or you're looking for evidence that the core business, which is the ad business, is improving because of their work in AI? Both of those things. And I think obviously, we're seeing the improvement in the core business already. And they talked about strength and growth into next year. There's already over 20 % growth expected into next year. And we've seen a pickup in not just user engagement, but also in pricing and in ad volume across a number of their different products. I think it is very important, though, that for products like Muse, that we start to see a monetization path.
32:01And that's where it's much less certain. And he did talk about typically Meta would focus on first drive users, first drive engagement, then monetization. But they've talked about moving much faster on Muse to be able to show that monetization at a much earlier stage. And that's going to be important if you expect investors to actually pay for the capex and they want to do an equity raise to show that they are delivering on returns from that Metamuse product.
32:33Ed Ludlow: The last sort of metadata point that I'm looking at is what came out in the filing this morning, which is the commitments, the spending commitments have gone to$825 billion. I think in July, it was about$700 billion. So they didn't guide for CapEx for 27 yet. But the direction of travel is pretty clear. Most people comfortable with that or that number is getting a bit scary with how big it is? I think, you know, the overall commitments, it will depend on how they show monetization. And I think that's very important that they continue to see leverage and strong revenue growth. They've also talked about, one thing that Meta doesn't have that Amazon and Google have is this third-party cloud business that gives them visibility on the CapEx that they have.
33:21And Meta just don't have that visibility to spend against. And I think that's why this is going to be an overhang on the stock through the capital, through this period, until we see the capital raise, until we get more colour on the size of that capex. And we really have to drive down compute efficiency because the unit economics on use are not attractive at present.
33:44Ed Ludlow: I think that's interesting because I'm already obviously thinking about next week. And if you say, OK, SpaceX is a NeoCloud and Meta is Meta, what did we learn about what might come in the balance of earnings season when it comes to the hyperscalar CapEx story? Well, we're seeing the strength that we're seeing from Meta. I think particularly even the early strength in Muse, the compute power needed. You need a lot of CPUs for that agentic portion. leveraging agentic that very much plays into amazon's gravitron and cpu product and we we've seen from the the leverage we had some disclosures from anthropic last week as well that that should also bode very well for the the tpu business at alphabet so we would expect to see continued strength in the cloud divisions of both amazon and alphabet next week and that continued build on their backlog especially from that vertically integrated setup that they have showing through to the supply chain that Meta doesn't have just now.
34:56Ed Ludlow: The story for quite a long time or for big portions of this year was to be long semiconductors and short software and there have been different parts of the year like the coming out of the summer where that may be changed, if not reversed. Alison, where's your thinking right now on the right place to be in the market? This is a great question. We can look back at past cycles, past compute waves, and we think we're moving out of that phase of just blindly buying physical infrastructure for AI, just buying all of the semiconductor, going long semis, short the software index. We think that's far too blunt.
35:37And going forward, investors are going to have to be very active because this volatility is here to stay. And also, they're going to have to be much more stock specific. There are lots of still really attractive opportunities in semis and physical infrastructure, but you're going to have to pick your points in terms of valuation and product cycles. And then the market, as this AI investment phase evolves, we're really looking for those companies that can drive applications and drive revenue growth from that. So we're focusing more and looking for those names and leading applications and leading platforms that can drive growth.
36:18We've seen some really strong numbers out of Spotify today. We're seeing a number of companies where we're seeing actual leverage on AI. We've seen some headcount reductions come through. And we think next year could be another attractive year of efficiency for tech. As we see companies, we're really looking at companies that can drive margin expansion in 2027.
36:42Ed Ludlow: The margin expansion is really important. Just on Spotify, it hit 788 million monthly users and basically had some fighting talk about how it's not going to get displaced by AI. the stock's up significantly as US listed shares. On margin expansion, I mean, where does the sort of free cash flow story for any given company or actually in aggregate feature for you? Like a lot of people seemed really calm when some absolute cash machines tipped into negative free cash flow, albeit temporarily or on a longer term basis. It's hard to gauge where the concern is on that. Yeah. And I think the important point is free cash flow is not disappearing.
37:20it is transferring around the ecosystem. So we have seen the hyperscalers continue to spend. But if you look at the free cash flow of NVIDIA, for example, we expect that that will continue to grow. And NVIDIA has been reinvesting that into the ecosystem. You know, very similarly, you know, we saw Apple back after the iPhone launch. They saw huge free cash flow growth and they had to invest that into the supply chain. And that's what we're seeing with NVIDIA. So, I think it's important investors don't think about free cash flow disappearing, but it's circulating in the system. And also, the point at which we expect to see some of these hyperscalers actually reach a free cash flow inflection point as CapEx actually peaks, which is likely to be into 2028.
38:10And then we expect to see some free cash flow inflection positively for them in the back half of 27. and 20 as well.
38:21Ed Ludlow: Alison Porter, Portfolio Manager at Janice Henderson. We covered a very broad range of subject matter. Appreciate your time. Thank you very much indeed. Now coming up, AI uncovers a major security flaw hiding in millions of websites. But could the same technology also give hackers an edge? Greg Brockman, OpenAI President, and Matthew Prince, Cloudflare CEO, joining us next. This is Bloomberg Tech.
38:47The 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.
39:28Go 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. Some people treat ChatGPT like some kind of smart search engine, and some use it to get work done. ChatGPT Work is a new way of working in ChatGPT that can take action across your apps and files, stay with a project for hours if needed, and turn a goal into finished work.
40:11It's designed to help you move from a chaotic starting point to a reviewable first version. So all the source materials, briefs, and scattered information that you have to grind through to turn into something useful can just become something useful. Put ChatGPT to work on your most ambitious ideas and projects. Get started at ChatGPT.com by selecting Work Mode, available on Plus and Pro plans. The world of business is constantly evolving. Every day, new challenges, new opportunities, new ways of working. Comcast Business keeps you totally in step with secure AI-backed networking built to power the way modern business gets things done today in more than 100 countries around the world.
40:55That's some serious muscle. And get this, Comcast business powers over 90 % of the Fortune 500 and millions of small businesses. Behind it all, network engineers, cybersecurity specialists, support teams, thousands of experts answering your call at 2 a.m. like it's 2 p.m. Always on, always ready. That's a lot of coffee. One partner powering how business gets done for companies of every size all around the globe. When you add it all up, no one does business like Comcast Business. Most meeting note takers get the easy words right and the important ones wrong. The client's name, the acronym, WhisperFlow NoteTaker is built to get the hard words right.
41:40It takes notes on any meeting platform, Zoom, Google Meet, Teams, even WhatsApp calls, keeps every meeting searchable and connects to Claude or Chat GPT so you can ask, what did we agree to last Tuesday? Free to try on Mac and Windows at whisperflow.ai. That's W-I-S-P-R flow dot AI.
42:00Ed Ludlow: AI helped uncover a potentially catastrophic vulnerability hiding for years in PHP. It's a programming language used to build and run millions of websites and web applications. OpenAI's technology helped researchers find the flaw. Cloudflare helped protect websites before that flaw was publicly disclosed. But AI could also help hackers exploit it faster. The concern is that the flaws could potentially let a hacker steal data or seize control of websites. Joining us now, OpenAI president Greg Brockman and CloudFare CEO Matthew Prince. I'm grateful to both of you for your time. Greg, I want to get specific here.
42:38Ed Ludlow: So what happened was Caliph used GPT 5.6 Sol to find a version of this bug back in August. And then OpenAI independently kind of found a more complete version in September. What is it that OpenAI's models actually did or do that allowed you to find something that I think researchers have not found for like years and years? First of all, thank you for having me. So you're exactly right. So Califf had a security researcher use our models very guided by that researcher's expertise and intuition. And a month later, we used our latest internal model that fully autonomously found this more complete version and very independently found the kind of full version of what was released.
43:25And the thing to realize about the AI models is that fundamentally they're empowering tools and that they're able to work across a lot more context than people are. I myself actually have found a security vulnerability way back in 2010 in an open source piece of software. I remember finding a small function that looked like it wasn't quite implemented correctly, but all of the usages of it had been protected. But then actually someone added some code later that didn't protect in quite the right way. And so it's this kind of sleuthing that you do as a person. But it's very hard to keep all that context in your head, to have expertise across many different disciplines, and to think about how the software might be used in practice.
44:05And so that capability, which is what makes AI so useful, that it means that it can help us solve problems. It's also something that helps us find flaws in the software that people have already written.
44:16Ed Ludlow: Greg, as a follow-up, a specific question I got for you is how much of the work in this case study was literally done by AI versus a researcher? So for the Califf side, I'd say, I don't know how to put an exact percentage on it, but the researcher and their deep expertise, let's call it 75%, 80%. whereas for the open AI internal version, that was really, we set up an environment and we took our most capable model with a lot of compute behind it and that autonomously was able to find the full exploit. Matthew, I want to go to you on probably the scarier bit, which was reading the blogs and the reporting.
44:53Ed Ludlow: It seems like hackers tried to exploit these flaws before they were made public, right? Which raises the question whether AI was what found them first. Do you see what I mean? Absolutely. And I think what we have to be clear about here is that AI is the real hero of this story. These flaws have existed for quite some time. They're extremely serious. And PHP, the underlying software that contained the flaw, is present and running on something like 70 % of the world's websites. So this was very significant, and it was a very significant flaw. It's hard to know whether hackers, nation states otherwise had access to this flaw that had been present for years.
45:36And I think that what OpenAI did with their tools and what the researchers did was extremely responsible. They found a latent bug that was there. They found something that may have been exploited for many years. They responsibly disclosed it to the industry and companies like Cloudflare. And we worked tirelessly behind the scenes to make sure that our customers were protected and that the Internet was protected before it got out. But it's clear it wasn't that AI created the bug. AI found the bug that was there. And it's incredibly important that we've now found it, fixed it, and the software is significantly more secure because of the work that OpenAI did.
46:15Ed Ludlow: But Greg, to the best of your knowledge, are there any examples of hackers using OpenAI's models to try and exploit this specific PHP flaw? So we believe that attackers are going to start using models broadly to try to exploit these flaws. We are monitoring on our side, watching for it. And I think we're starting to see attempts or we have seen attempts in the past, but not at massive scale yet. And I believe that we're going to see just a ramp up. I think that Cloudflare has reported seeing this kind of very steep ramp of attempted exploitation. So I think that we should expect attackers to try.
46:56But as Matthew said, that we have put in a lot of effort here to do responsible disclosure, to repair the Internet and to make this software more secure.
47:05Ed Ludlow: Matthew, same question to you. sort of knowledge or awareness of any specific incidents where open AI models have been used or an attempt to be made by hackers to use them. But also kind of give me the backstory of when Cloudflare came into this, when you got involved. So first of all, once the exploit is known, AI doesn't have to be used at all. You can actually, in this case, with a couple of requests to a web server, take over that web server. So once we know what the problem is, then that can be exploited without using AI at all. And hackers right now are exploiting this flaw at just incredible rates.
47:43Literally, we're seeing millions of attempted attacks using this flaw now across all of our network, you know, every single minute of every single day. So the flaw is now known. And if you are running PHP on your servers, you need to update PHP to the latest version. It's free. You can get it updated immediately. And that's the only thing that's going to keep you safe from this. What I think, again, was incredibly responsible of the OpenAI team, and I think most of the frontier labs are working with companies like Cloudflare in this way, is that when they found this flaw a couple months ago, they reached out to us and they said, we've found this.
48:25They disclosed it to us in a confidential way. We were able to then work with our team to make sure that we had put protections in place so that when this was actually released, as much of the Internet was protected as possible. And they didn't just work with Cloudflare. They worked with WordPress, with the PHP community, with anyone that could actually make as much effort to protect this. And again, I think it's really important to say here, AI is the hero here. This is a latent flaw that has been in the code for a long time. It may have been being exposed by and used by nation states or other hackers to do all kinds of bad things.
49:02But now that it's been exposed, we've actually fixed it. What I'm optimistic about is, yeah, there may be times when, you know, bad guys use AI to find things. But right now we've got the good guys who have more resources, have access to great tools like the ones provided by OpenAI and the ones provided by companies like Anthropic. and we're going through code and we're going to find a bunch of flaws. We're going to find a bunch of bugs. But I actually think the Internet is getting more secure because of the work that companies like OpenAI are doing.
49:31Ed Ludlow: Greg, you wrote quite a pensive post on X, sort of recalling your own history with PHP as a programming language. One of the ideas that you talked about was suggesting that the governments or I think you call them authorized defenders, but they scan the Internet for vulnerable systems and in some cases just remotely patch them. And I think that raises a lot of questions about permission, like who decides that that should be the case and where the line gets drawn. So I think that we are, you should look at PHP as one example of the kinds of flaws that AI is now able to find. And we're in this world where there's what we call a defender's window, where frontier models are ahead of what's broadly available.
50:13And so that means that the labs, those who have access to these cutting edge capabilities, can find vulnerabilities, can peer into the future before anyone else does. And so that means if you are a company and you secure yourself using the frontier models, you will be more secure and able to stay ahead of what is broadly available and what attackers can do. And so I think that defender's window is a very important concept for every company, for every defender to really think through. But it's also very clear we're going to be finding lots and lots of flaws that are there in the software that have been there for many years that are tech debt that have piled up as people have built software on top of, you know, just many, many years worth of accumulated tech debt.
50:55And so to me, the question is, how does the Internet stay secure? And there is precedent with, for example, as botnets became prevalent, having the FBI get quarters to clear out malware from botnets because that's something that's important. It's not just the site owner who is affected if they get attacked. It's really the whole Internet becomes less robust and secure if there isn't someone who's trying to make sure that patches are applied or that if there is sort of malicious activity, that there is some sort of way of notifying site owners and making sure that improvements are made. And so I think that the core question we have, I don't think that we have the answers yet, but I think that we're really asking the question is, as the scale of vulnerabilities found increases, and it's harder and harder for anyone to make sure that they're taking all of the patches, how does this whole system need to change?
51:51And so if you look at PHP, we had two months worth of very manual process. We were on the phone with many of our partners, the government to really try to land this well. We see other vulnerabilities coming. We've been working on several of them as well. And I think that the system of making sure that we stay ahead, that defenders are advantaged, and that there is some sort of process, some sort of societal process for keeping the Internet secure, that's the question that we think that the people should be asking.
52:20Ed Ludlow: Greg, before I let you go, I've been using Dots a lot. Alongside Muse, GrokBot, I kind of put this question to Sam the other day, trying to work out how to make an agent useful to me. Is there any updated data you can give me on how Dots has gone from an adoption perspective and a user perspective? It's really hockey-sticking. I mean, it's really exciting to see all the usages. And actually, I think that there's so many surprising things across work and also your personal life. And even, for example, for protecting people, one of my friends told me that her dot noticed a phishing email in her inbox said, hey, do not open this email.
53:04It's a phishing attempt. And her admin also has access to her email and it also pinged her admin saying, do not open this email. And so I think that we're finding this proactive form factor of an AI that's looking out for you in your personal life, in your work life, to be something that just increases people's, helps people save money, is able to increase people's productivity. And it's something you could not get a person to do, right, to be looking out for you constantly. So I think that there's just so much potential that's being unlocked. And I think that we can make people more productive, make work more joyful and also help people just live more fulfilling lives.
53:43Ed Ludlow: Greg Brockman of OpenAI, Matthew Prince, CEO of Cloudflare, Cloudflare Key to Protecting a lot of what's happening in Internet infrastructure. Thank you both very much. That does it for this edition of Bloomberg Tech. Look, it's been an hour dominated by earnings stories. in the case of Meta and Tesla and IBM, AI spending is a big theme. But in the market, I'd say the kind of prevailing sentiment is AI concern or AI risk that's out there in equity markets. Also a fascinating discussion of a company keeping its founder CEO in place amid a big political story here in California. Recap all of that on the podcast.
54:20Ed Ludlow: You know where to find it online on iHeart, Spotify, Apple, and of course, on all the Bloomberg platforms. Thanks for being with us. We're almost at the end of the week here in San Francisco. This is Bloomberg Tech.
55:07This is Robert Smith from Business History. If you're listening to this, there's a good chance you're a small business owner. And like every small business owner, you started with a dream to do what you love and watch it grow. What you probably didn't dream about? Keeping up with cyber threats. That's where MasterCard can help, with access to tools that help identify cyber threats to better protect your business. Building your dream business? Priceless. For cybersecurity in a changing world, there's MasterCard. Learn more at MasterCard.com slash smallbusiness. This episode is brought to you by Choiceology, an original podcast from Charles Schwab.
55:43Hosted by Katie Milkman, an award-winning behavioral scientist and author of the best-selling book, How to Change, Choiceology is a show about the psychology and economics behind our decisions. Hear true stories from Nobel laureates, authors, athletes, and everyday people about why we make the choices we do and how to make better ones to help avoid costly mistakes. Listen to Choiceology at schwab.com slash podcast or wherever you listen. When it comes to my policy, Farmers gives it to me straight, so there's no surprises. It's your coverage. You should never be confused. You hear that? An insurance provider with a conscience.
56:26We're a choir, but yes, we have that too. Straightforward coverage you can literally understand. That's what you get when you have Farmers. Bum, bum, bum, bum, bum, bum. Learn more at Farmers.com. Underwritten by Farmers Truck or Fire Insurance Exchanges or Affiliate. Products not available in every state.
Bloomberg’s Ed Ludlow breaks down Meta's plans to boost AI spending and release new Class C shares to ensure CEO Mark Zuckerberg stays in California if a wealth tax is passed. Plus, Tesla reported strong profit, but a big cut to its spending plans is raising questions about Elon Musk's AI, robotaxi and robotics goals for the company as it weighs a potential merger with SpaceX. And OpenAI's models helped find a potentially catastrophic software vulnerability. OpenAI President Greg Brockman and Cloudflare CEO Matthew Prince explain how AI is enabling both defenders and hackers.
See omnystudio.com/listener for privacy information.