The Six Five Pod | EP 270: Silicon Strategies: Intel's 14A Gambit and the Future of Chip Manufacturing

On this episode of The Six Five Pod, hosts Patrick Moorhead and Daniel Newman dive into the latest tech news and earnings reports. They discuss Intel's foundry strategy, NVIDIA's chip sales to China, and recent earnings from major tech companies like Amazon, Microsoft, Meta, and Apple.  Get their analysis of market trends, geopolitical factors affecting the tech sector, and the competitive landscape among industry giants. This week’s handpicked topics include:

‍

1. Discussion of the EU's 40 billion Euro AI chip purchase agreement with the U.S.: A quick analysis of the practicality and implications of this deal. (The Decode)
2. Palo Alto Networks Acquisition of CyberArk: A look at their potential $25 billion acquisition of CyberArk and the strategic importance of identity management in cybersecurity. (The Decode)
3. NVIDIA Updates: Concerns about potential backdoors in NVIDIA’s H20 chips and an analysis of the insatiable global demand for NVIDIA’s AI chips. (The Decode)
4. Microsoft & OpenAI’s Ongoing Relationship Challenges: Commentary on Satya Nadella & Sam Altman’s respective strategies (The Decode)
5. The Flip - Intel's 14A Foundry Strategy: A simulated debate on Intel's approach to 14A Foundry development, including a discussion of geopolitical implications and competition in the semiconductor industry. 
6. Bulls & Bears Tech Earnings Roundup - Analysis of recent earnings reports from major tech companies:

- Amazon's impressive growth and future potential

- Microsoft's strong performance, particularly in Azure

- Meta's success in leveraging AI for platform stickiness

- Apple's steady performance and potential AI developments

- Arm's strategic shift towards chip production

- Qualcomm's challenges in market perception despite solid performance

7. Market Reactions and Future Outlook: Market responses to earnings reports and strategic moves, and speculation on future developments in AI, robotics, and data center technologies.

‍

For a deeper dive into each topic, please click on the links above. Be sure to subscribe to The Six Five Pod so you never miss an episode.

‍

Or listen to the audio here:

‍

Disclaimer: The Six Five Pod is for information and entertainment purposes only. Over the course of this webcast, we may talk about companies that are publicly traded and we may even reference that fact and their equity share price, but please do not take anything that we say as a recommendation about what you should do with your investment dollars. We are not investment advisors and we ask that you do not treat us as such.

‍

Daniel Newman: You know, Satya, I'm over you, I'm under you, I'm around you. You know. You know, Sam wakes up in the morning, he looks over and he just sees Satya. It doesn't matter who's laying next to him. It's just Satya is there, just staring at him, eyes wide open. 

‍

Patrick Moorhead: What a great meme.

Patrick Moorhead: Welcome Back to the 65 pod. This is the Pat and Dan Show. We're at episode 270. It's the end of the week. I know it's Monday for all of you, but it is the best hour of the week for me. Daniel, how are you doing, my friend? 

‍

Daniel Newman: As good as I look. I got raccoon eyes. It's a little red in here. Been doing a lot of tanning lately, Pat. You know. 

‍

Patrick Moorhead: I just. I'm just glad you get out there into the sun. You know, you get to work like two or three hours a day and. And then, you know, lower your stress. I get it. 

‍

Daniel Newman: Well, I. I run a culture of performance, so the performance here is we all compete to see who can take the most PTO. I've got some real winners. Some actually take more days off than they work. And. And we're still crushing it. So, you know, around here, it's all about. It's all about the culture. And so I've been taking a lot of time off, a lot of time at the pool, clearly wearing my shades. That's why I'm so red. But, you know, listen, you know, I come back for the one hour each week that really matters, and that one hour is with you here on this podcast. Best hour, hopefully for all of you as well. 

‍

Patrick Moorhead: There we go. But Dan, you know, I love the sarcasm. It's pretty awesome. And I think. I think at one point I was texting you, hey, bestie, where are you? I need somebody to be miserable with. 

‍

Daniel Newman: It did happen. And then I called you back and said, now I need you. 

‍

Patrick Moorhead: Yes, I know. No, we are. Some people confuse us as. I mean, it's weird. Like, we're so different in how we think and what we do, but we are aligned on some certain, certain entities. One of these is crushing it. Working hard, aggressiveness, losing is unacceptable and taken personally. But how’s the workout going? Dan, you had a bum shoulder when, when you, when you were out. Are you back? Are you fully back? 

‍

Daniel Newman: I am as back as I can be. I am fully committed to trying to hurt myself again. So all in a series of a single week. I, you know, tried to play soccer with my son and I fell on my shoulder. That created some long term pain, but you know, that's okay. And then I tried to play again. I was kind of straddling the shoulder and I was running around playing and I pulled a hamstring. This by the way is part of just being old. Like I, I am pretty consistent with fitness. You know, I'm on, on it but like just getting old sucks. I stick Pat. I shot a little BPC in there. BPC is a good thing. Everybody check it out. 157. It's the best. Add TV, it's even better. I mean it's not, not, not health advice. So do your own research. And so I got better pretty quickly today. I sprinted again. So I actually was able to do sprints up at 4:00 AM staring at the ceiling and I was in the gym by 5:00, doing my sprints by six and back in my chair by seven plowing some protein shakes and green tea. So you know, the health thing's going a lot better than other things right now. I gotta be candid. Like the whole thing is, is in line. And thank you by the way for the motivation because like a year ago I was, I was, I was a pretty, you know, I was not as good looking. Now you know. What did you call me yesterday? Unrecognizable. You told me. 

‍

Patrick Moorhead: I. Yeah, yeah, literally you sent me that. The naked picture of you. 

‍

Daniel Newman: I wasn't fully. Like I'm just throwing out there who already suspicious of things like it wasn't fully. There was maybe some abs showing though. There might have been some, some abs or flabs. You'll have to be the verdict of whether it was what. 

‍

Patrick Moorhead: Well, hey, I'm glad you asked about me though. 

‍

Daniel Newman: People are going to take it seriously. They're going to be like Dan and Pat are sending each other naked photos. 

‍

Patrick Moorhead: It's like, well, well maybe, maybe that's the what we want to portray out there. So anyways, so I am, I am. I really do appreciate you asking me how I'm doing. 

‍

Daniel Newman: How you doing, Pat? 

‍

Patrick Moorhead: Yeah. 

‍

Daniel Newman: Wait, wait. We were doing role reverse. It's all about me. 

‍

Patrick Moorhead: No, no, I know, but until it's about me. 

‍

Daniel Newman: How's your fitness, Pat? Are you jacked? Are you totally jacked again? 

‍

Patrick Moorhead: Well, I'm seven days in from surgery and I wasn't supposed to to lift anything for, for two weeks. It still aches, I gotta admit. But I stopped taking even acetaminophen three days ago. My HRV is absolutely in the dumps as a percentage to normal. It was down to 23% and which is like, don't move, don't do anything. I was on a business trip in Kansas City when that thing hit. Right now, you know, I think I'm, oh, I'm at 50%, which is, you know, kind of iffy. But I was rocking the 100% for a while. My physiological stress was super high as, as, as I was healing. But no, you know, I'm back and I'm gonna hit the, I'm gonna hit the treadmill on, on an incline, do what I can do and pick myself up. And I, I see my doctor a week from today and we're going to talk about whether I can, what kind of weights I can start hitting. So, yeah, I'm not back, but I feel incredibly improved here. So this is a tech show. Let's dive into the tech. So we've got some great things we're going to Decode for you, some big news items that, that came out. We're going to hit The Flip where we essentially argue the polar opposite ends of very nuanced issues that don't have polar opposite answers. And then we're going to hit big market week. I mean, you know, if it's not just the earnings, but it's tariffs, if it's continued discussions about interest, interest rates. So we're gonna, we're gonna have to see. So let's, let's dive into The Decode. All right, Dan. EU plans a $40 billion Euro AI chip purchase in U.S. trade agreement. Oh, wait a second. I didn't think the U.S. even made chips. What's going on here? 

‍

Daniel Newman: Yeah, can I just really quickly say, because you know, on it's Friday the 1st of August, so welcome to the new month. Trump came out with a big old list of tariffs last night, as if this was surprising. And the market's dumping. You've got, I just want to call a couple things out that I don't know if we're going to fully cover. You've got Fed Waller, Fed board member Waller and and Jay Powell fighting in public. The jobs data sucks. So just important point, important to point that out, but the market as we are speaking right now is doing a little nose dive. Is it a buy opportunity? Anyways. But last week, right? As we headed into this week, there was this year Europe deal. There's supposedly, I Think a Taiwan deal got done, a Japan deal was nearing in India deal. At least a percentage with India has been disclosed. But, you know, one of the big things about the agreement was this big commitment to buy chips. And you know, this was kind of interesting to me because it's like were they not going to buy them anyway? Is this just kind of padding the victory here? Like, you know that Europe's got more like data stringency than any other part of the, of the world. They, of course, are going to have to set up sovereign clouds. We've heard Jensen talk about massive deals in Europe where they were going to buy a ton of NVIDIA GPUs, because every country, in order to stay within the data protection rights that they've created and the privacy rights that they've created, will need to have complete control and sovereignty of their infrastructure in the region. So this is going to lead to them buying chips, this is going to lead to their businesses buying chips. This is going to lead to the cloud providers adding zones and adding more chips. So I think it's great. I think to some extent though, this was a little bit of a padding on a victory lap that was already happening on something that, you know, maybe it would have been more odd if they said Europe decides not to buy chips and the fact that they're buying them from the U.S. it's as if there's another option. There is no other option. Even China doesn't want their own chips right now. And yes, I'm being a little facetious, it won't actually end up that way. But there is a reason that Jensen had to expand his H20 manufacturing because China wants a heck of a lot of H20 chips because they really want NVIDIA. 

‍

Patrick Moorhead: Yeah. I can't believe this is our first topic, but it is. 

‍

Daniel Newman: That's why I let it. That's why I gave the little leader. Because there's a couple other things going on. I think there was a. The other stuff seems to be ripping today, you know. 

‍

Patrick Moorhead: Yeah, this one looks like nonsense to me. The, the only thing that. And we had a nonsense. We had a nonsense topic last week. I forget what it was, but as I was going through, I'm like, why is this even a topic anyways? 

‍

Daniel Newman: Nonsense. You know me, by the way, the. 

‍

Patrick Moorhead: The only thing I could come up with if I'm inventing something, Daniel, was you have Intel, who's manufacturing chip server chips in the United States, and then you had AMD who is manufacturing their chips in Taiwan. That was the only thing I could come up with, right. There's no GPUs that are done in the United States. There's no high speed networking chips that are, are done in the United States. There was a lot of analog stuff, right. Done at let's say Global Foundries. There's a lot of memory that that Micron does, but like they weren't going to buy it in the first place anyways. 

‍

Daniel Newman: This is nonsense. 

‍

Patrick Moorhead: Let's, let's move on. 

‍

Daniel Newman: But I mean it was a nice padding on the victory lap, right? I mean you and I love a victory lap. It's like they're going to bite chips. It's like no s*** Yeah, everybody's buying chips. Have you seen this week? Anyway, exactly. 

‍

Patrick Moorhead: By the way that the article came out in investing investing.com. so okay, let's move to a real topic here that actually has meat and that is that Palo Alto is making a play for CyberArk. A $25 billion deal. You know, you and I sat down with the CEO of Palo Alto back at, I actually forget which, which conference that was in was San Francisco and I got to tell you he, he operates at a different clock speed and he is dynamic and he wants to rule the world. We also met with their founder uh, in uh, Spain at MWC and really got got his feedback. But essentially if it's not cloud based security then it's shite. Okay, this is what Palo Alto does. The control plane being in the cloud. The thesis being things move so quickly in the age of AI. You have to be doing this in the cloud. But what does CyberArk bring? Well, it brings identity management, it's called PAM, Privileged Access Management and identity management to their portfolio of end-to-end cybersecurity, and you know it's, it's funny Dan, you and I were asking the founder I think at MWC, well, well what about all this security stuff that's maybe on device and he's still stuck to his gun. And we talked a little bit about identity management here. So anyways, this would, this would make Palo Alto one of the biggest secure. Well the biggest security pure plays out there. But bigger, even bigger than what Microsoft brings to the brings the table. 

‍

Daniel Newman: Yeah, I mean you've got an industry with tons of, you know, fragmentation, not necessarily a clear winner. It's a fast grower. It's one of the biggest adjacencies to AI itself. As AI scales, security will scale. There's kind of this RSA type of part and of course the quantum part that's going to come into play Quantum encryption because in the next few years we're going to probably break RSA and then Bitcoin will be vulnerable. Our good friend Niccolo will tell you all about that. And of course that's what he wants at IonQ, for them to break it so they can sell more quantum networking and security. But anyways, the, the netpad is, it's an opportunity. I mean look, this industry just does not have a dominant player. It's, it's Microsoft. Microsoft is the dominant player, but it's like there's not a pure play dominant player. Palo Alto is pretty much as close as it comes. And you know, as you know he told us when we spent time with Nakash is that, you know, he's very accretive. And so you know, we heard last week it was sentinel 1. A week later it was CyberArk. By the way, CyberArk was number seven on the Futurum Cyber 7 and Palo Alto is number three. And so coming together it, you know, we may have a new opportunity at least number two to move ahead of CrowdStrike. But this was an area that Palo doesn't do. So it's almost purely accretive to their business and so makes a ton of sense. By the way, the street absolutely beat the crap out of Palo Alto stock for this. They lost more market cap than the deal size in terms of the, the share. But the CyberArk stock people loved it. So you know, we know that that's generally kind of the reaction is not usually this harsh. But having said that, I think Palo knows what it's doing. I think it's a good move and I think sort of consolidating for scale is important right now because you know, with Microsoft and obviously Google with Wiz, these big companies are not going to just leave this market untapped. They are going to try to bring it all in house and make it part of the, the, the Amazon and the Google and the Microsoft cloud. Because why would you want to have to have a point solution? So Palo needs to be so entrenched, so inundated, so diversified and so specialized that people are like no, we need Microsoft Azure for all this, but we want Palo for all that. And so he needs to keep building scale. I think it's a, you know, it's a good deal. I just, I actually think if you like callow after that beat down on the stock, maybe now is a good time to enter because it definitely took it but it still trades it. I know we're not, it's, we're not doing buys and you know me, I love talking about this crap, but it's like it still actually trades at a really high premium. So just kind of watch for that. But I like security. 

‍

Patrick Moorhead: Yeah, this whole consolidation mode really started off with Cisco, and then it went to Microsoft and now Palo Alto Networks and fundamentally Enterprise IT. What they found themselves in is they realize that even if you're, you know, one off pure play security company that does one thing well, the time it takes you to integrate that into your end-to-end security portfolio, by the time, you know, you got it to work, it was too late. And it didn't end up being the best security. So Enterprise is looking for less valid vendors to deal with. Hey, let's move into. I mean, my gosh, we're 16 minutes into the podcast and we haven't even talked about NVIDIA yet. Lots. 

‍

Daniel Newman: Three times, dude. I mean. 

‍

Patrick Moorhead: Oh, okay. 

‍

Daniel Newman: Remember I kept mentioning Jensen said this because I'm such a dork, I can't help. And I gotta start quoting you more. 

‍

Patrick Moorhead: Oh, I know you're a. 

‍

Daniel Newman: You were on TV yesterday. Didn't I see you? 

‍

Patrick Moorhead: Yeah, you're an NVIDIA pumper. I get it. And it's such a hard…

‍

Daniel Newman: Did you just call me plump? It's the way I'm sitting that makes me look like I have boobs. 

‍

Patrick Moorhead: It's pump, not plump. 

‍

Daniel Newman: Pump. Okay, plump. 

‍

Patrick Moorhead: All right, let's go to a topic literally everybody wants to talk about. And that's NVIDIA, right? We got 20, you know, 300,000 H20 chips. But you know, we've got China calling NVIDIA on the carpet. What's going on here? I mean, is this really bad news? Good news? 

‍

Daniel Newman: FUD. Another stupid topic that isn't stupid. Like, because I kind of mentioned this on the E.U. stuff. Like, everyone's buying. H20 demand is through the roof. Huawei, you know, is brute forcing it, but, you know, is still behind. But this could be a little bit of China pulling a little lever on national security. Pulling a little lever on, you know, they're actually seeing that like the Ascend the Matrix isn't quite winning, like all this. It's like they wanted access, but it's like, I don't know that they wanted it to be quite this good. It's like this is better than we thought because it's all, in my opinion, you know, just like China's doing with EVs. It's all about a bridge to China. Like, they want access to what they need to keep moving, but they also kind of ultimately, eventually, will want people to move to their stuff. And even different than like handsets with Apple in there and, and cars, this is a bigger issue for them long term that, that they're going to build this dependence. So we're diffusing our tech. Their researchers love it. It's not just H20s, but we did talk about kind of what are some of the, the things that could be required that cause issues. You know, I kind of tweeted something along the lines. You and I have said the same thing about like, kind of know your customer. There could be a time down the line where China doesn't really want anyone to know where these H20s or any other chips from NVIDIA or AMD are. We obviously know that there's still some intent within this AI Strategy from the White House to not feed China our best technology, which again, the H20 is not our best technology. But let's be very clear, H20 is not the only NVIDIA chip getting into China right now. So there's a lot going on with this. I think this is just the continued game of, you know, chess that's being played with everything from, you know, you know, the critical mineral and, you know, APIs for Pharmacy. And of course, AI is our lever. So AI is our lever. They want to obviously make sure that they keep us in check wherever they can. 

‍

Patrick Moorhead: I got two things to somewhat add to this. So NVIDIA having to build 300,000 extra chips on top of the, what, the million that were supposedly there before says that even detuned NVIDIA chips are in higher demand than Huawei Ascend chips. I think there's 2 to 3 million of those Ascend chips that are, that are available. But I, I just think it's, it's, it's amazing. Just the pull for this type of chip and then, you know, on the back door, I, I just think this is complete and utter nonsense because if this were the case, the, the U.S. government would have been allowing H20s to be shipped over by the truckloads at a 50% discount paid by the CIA to be able to do this. And you know, maybe there, there is a lot of software that NVIDIA does do, like Dynamo, right? Like data management software, like that does update like any other reasonable software. And oh, oh, oh, guess what? Those GPU blocks also have firmware that need to be updated. So I just, I don't know if there was a back door, we would have been sending them over by the truckload. Therefore, my logic, I don't think there, there's a back door. 

‍

Daniel Newman: Well, just you and I agree. The insatiable demand is indicative of just how much absolute BS and FUD were being fed in different commentary, press, media. And I'm just glad guys like you are here to, to dispel stupidity, you know, because if they weren't, there have been a lot of people probably on the wrong side of, of a number of various risky positions, including shorting NVIDIA clowns. It's just a terrible idea. 

‍

Patrick Moorhead: I don't know bestie it's not just me. So this is what The Decode is about. But you're as big a or bigger truth teller as, as I. 

‍

Daniel Newman: I just – you know, someone told me, they said to me, Pat, they said, I like your Twitter now. They said, you know, you're like, that's the real you. And I said, yeah, you know, I brought the, I've brought the snark. I said, I just hit the point. I just don't care anymore. Like I used to really care. I don't. I. I'm thankful for that. Maybe it's like you hit a certain age. Like 44 was my moment where I just, you know, I had some guy because I showed a picture of me like training and staring at myself and I said, my own competition. The guys like sent a picture. It says, you're gay. 

‍

Patrick Moorhead: That's so good. 

‍

Daniel Newman: So they say, I don't know. But like, you know, I'm just, I used to think like, I don't want anyone to say something mean to me. It's like, you know the stuff they say in public. I just can't imagine what they say when I'm not around. 

‍

Patrick Moorhead: I mean, that stuff used to bother you and I'm so glad it doesn't anymore. And I'll admit it, it gets me. It's depending on everything else that's going on, right? Like if everything sucks, business, personal, physical, and then that hits. 

‍

Daniel Newman: It's like extra sensitivity. 

‍

Patrick Moorhead: Exactly. Hey, let's move into our next topic here if we can. Although, I mean, I would camp out on this all day long. Groq with a Q that both you and I are on the cap table early on this one is going for another $600 million in funding that supposedly will put the valuation to about $6 billion. By the way, a little background. So first of all, Groq, their founder, Jonathan Ross, is the, the essentially, one of the founders of the TPU that is arguably at Google, the most successful non-GPU AI accelerator on. On the planet. That is that has been brought together and you know, they have data centers in the U.S. and Canada and Saudi Arabia and now Europe I think. And what they did is which takes a ton of money is they're not just building the chips, they had to build an entire entire solution. And then they went to “as a service,” basically “Groq as a service” and they're doing this all currently with a 14 nanometer chip that's fabbed out of, out of Global Foundries. And that's compared to, you know, NVIDIA, sorry AMD that are, you know, 444 nanometer ish on the silicon. So that with the performance and the throughput that they're getting out of that and the power is pretty impressive. Now you have to string together a lot of Groq servers to be able to get the performance to do a large model. Right. Because it has a minimal amount, I think it's 200 - 300 megabytes of SRAM that's actually on the chip and you have to fit the entire model size into that. Now the way you program it is do different weightings, which means you don't have to have as much as memory as you would in HBM sitting on a, on a, on a GPU, but it is a more efficient approach. So, very interesting. You know, you and I have debated the whole ASIC versus GPU what happens to this. If you look at the COAS loadings that are are rumored out there, folks like Morgan Stanley and, and JPM, you know, it still looks like, let me see, AMD I think is around 11% market share. The TPU is at 15% market share. And then kind of the odds and ends of kind of the SambaNovas, the Cerebras and the Groq’s combined I think are around what, 5% – 2% to 5%. And then NVIDIA has everything else. So yeah, very interesting moment here, particularly as we're moving, not moving, still doing a ton of training, but as we move to actually having to run the workloads. 

‍

Daniel Newman: Yeah, I don't have a lot to add here. Got a little lucky on this one, on a personal anecdote because you know what they were when we actually came in and worked and got on the cap table was a different thing. They thought they were building a custom training SL inference chip. Quickly discovered that that wasn't the right path. Went all in on inference. They won some great deals in Canada, in the Middle East. Really became a full on cloud. They're basically an inference cloud, you know, is sort of a Coreweave kind of thing. But just for inference and have been very kind of like selective specialized for those that need tons of horsepower and tons of low latency. Definitely a fan of Jonathan. We've. He's a multi-timer on Six Five so we've had him a number of times. And their, their new CEO Sonny Madra is great. He's a guy I try. I think you probably track him too on social. You probably met him by the way when you were. 

‍

Patrick Moorhead: Yeah, I met him. I met him for the first time in Washington D.C. at the AI event. 

‍

Daniel Newman: He definitely like when they made that acquisition he came in, everything changed. It was a really good strategy like when he and Jonathan got together because I think they bring a really complimentary skill set. He's. He's a killer though. Like he's absolutely. 

‍

Patrick Moorhead: He is an absolute killer and he's been very good. I, I totally agree. 

‍

Daniel Newman: So good stuff. Let's go 6x Pat. It's a six bagger. 

‍

Patrick Moorhead: Let's go six banger. 

‍

Daniel Newman: At this point when you're sad about, you know you're. You're pulling your pockets out and they're empty and remember you got a six bagger in Groq. 

‍

Patrick Moorhead: I don't know. Last work for equity deal I did, I got a check for $3,400 that I split with somebody. 

‍

Daniel Newman: Yeah, it was pretty good. I'm sure that, I'm sure that you probably put the, you know, was that about a 10 minute block of time that you would have. Should have committed to on the average. 

‍

Patrick Moorhead: Of what you charge per hour, 45 minutes. So no, I do. I do one off consulting between $2k and $5k. So yeah, it was pretty. It's pretty sad. By the way, it's better than me. I mean I invested $225,000 in a SPAC that was in that horrible in between period between actually coming out and, and not happening at all. And I lost a quarter of a million dollars in that. So. Yeah. 

‍

Daniel Newman: Well you know Pat, I guess better to only have lost that than more. I mean. 

‍

Patrick Moorhead: Oh yeah, let's talk about. 

‍

Daniel Newman: I wanted in on that deal. So I'm really glad you didn't make any intros right away know. 

‍

Patrick Moorhead: Well let's not forget about. I think I lost $150 grand on a. On a. On an. 

‍

Daniel Newman: What were those things? Token. 

‍

Patrick Moorhead: I don't even know. Some you know basically tokens art nonsense. 

‍

Daniel Newman: Anyways, you were gonna sell what was it like? The, the monkey. The ape. Bored Ape. You were going to be in the business of Bored Apes. 

‍

Patrick Moorhead: Exactly. I thought it was like a sure win situation. 

‍

Daniel Newman: Wasn’t that when every livestreamer became was a, was a, you know, some sort of crypto or what do they call the, you know, non fungible NFTs. They all became NFT people and now they're all Bitcoin “laser eye” people or whatever. 

‍

Patrick Moorhead: Yeah. Anyways, this is why you never ever take anything I say as investment advice. You do not want to be me, okay? Because all I do is just lose money. Terrible investor nonsense. Hey, great guy, Microsoft news. A lot of stuff going on, you know, a lot of talks about OpenAI agreements. Are they going to let them turn into a full for profit? Right. We've got Copilot Mode and Edge and oh, Microsoft, SharePoint got hijacked. 

‍

Daniel Newman: Gosh, Microsoft has so much leverage on OpenAI ridiculous. Like you know, Satya, I'm over you, I'm under you, I'm around you, you know. You know, Sam wakes up in the morning, he looks over and he just sees Satya. It doesn't matter who's next to him, it's just Satya is there just staring at him, eyes wide open. 

‍

Patrick Moorhead: What a great meme. What a great. It's probably a meme, isn't it? 

‍

Daniel Newman: You see my funny meme, you know that meme of the guy like with the girl looking back and kind of whistling. It was really popular, you know, like he looks back and it's, you know. Anyways, it was all over the Internet. I shared it. Here, let me see if I can, I can, I'll show it in a minute. But, but I shared about Apple like all the Apple investors looking back at, at the Meta investors yesterday. Like, but. 

‍

Patrick Moorhead: Oh no, no, I saw that. Yeah, yeah, yeah, yeah. 

‍

Daniel Newman: This one, this one, it was really popular in its time. But like so I put a little hat, I put the Apple ticker on the, on the girlfriend and the meta ticker on the one he was checking out. But anyways that was a lot of fun. But look, I mean Satya needs to get control. There's this AGI moment which is sort of indiscernible of what that even means where supposedly OpenAI kind of, Microsoft loses control. And obviously with all the money being poured in and how fast Sam is going and how ambitious slash diabolical he is, I think Satya wants to make sure he reels him in and gets control and doesn't lose because Microsoft is a great growth and we'll talk about that later. But like Microsoft also this interesting just position like Gemini and Google and TPU, they built this whole stack. They're like self sufficient and if NVIDIA cut them at the knees tomorrow, their business impact would be different. Amazon's going down a very similar path to them. Microsoft has this great growth rate, but they are sort of like kind of doing the best job of selling other people's innovation. Meaning they're selling Copilot, which is based on OpenAI. They're selling mostly they're the biggest CapEx spenders on, on Blackwell right now and selling NVIDIA, they built a ton of capacity which by the part of the reason AWS was down was capacity. They're down unconstrained on capacity and energy and chips. So you know, I'll touch this one, I'll pass it back to you for the others, Pat. But like I think Satya's got to work the best deal to make sure that he doesn't lose access or the proprietary nature of all these solutions they've built. Because as of right now they don't have a model and you know, I mean they have some small models but they don't have a big foundational frontier type of model. And so this is super, this is, this is like critical that they get this right. But it also like they have the $10 billion in and it's worth like $160,000,000,000 now. So Satya has like 16 times the revenue in the investment while also getting access to all this tech, it'll probably double again. So once he lets the chains free, I mean figma, $60 billion OpenAI would probably be like a trillion out the gate. Which by the way, I know we didn't talk about Figma, but that is an absolute nonsense valuation for a company that cursor could basically do the same thing now, like build me a, you know, like cursor on an LLM. I don't know, I love that these things happen. I just guess I'm being bitter because I'm not in on it. 

‍

Patrick Moorhead: So there we go. Instead we're, I'm buying crypto monkeys and SPACs. SPACs that never happened. 

‍

Daniel Newman: What a terrible bunch of investments you made. I'm kidding. 

‍

Patrick Moorhead: God, I can't believe we have Bulls and Bears. Anyways, we'll get there. Hey, the only thing I want to add on this, on the Microsoft thing is SharePoint and Chinese hackers. So a couple fact checking elements I want to do on here. So first of all, cloud-based SharePoint in Microsoft 365 wasn't impacted at all. Okay. It's the on-premises SharePoint that was and namely the unpatched ones, older ones that hadn't been updated in a long time were the ones that, that, that, that were targeted. And you know, within a few days of, you know, these breaches coming out, Microsoft issued a patch that's in accordance with its new SFI protocols and net-net Microsoft met its 90-day SLA. Right. The issue was identified May 16th. Patch was available three weeks later. And now after the patches, Microsoft identified new exploits with a new patch came out on July 19. And even in the interim, a lot of these anti malwares, 85% of these were blocked by Microsoft technologies. So net-net is patch your systems immediately if they're on-prem. Don't wait. If you somehow for some reason can't find a way to operationalize this, you really should consider moving this workload to the cloud. But also understanding the, the, the pros and cons of, of, of doing that. I think this is another instance where the press kind of grabbed onto a story and ran in a certain, in a certain direction. But yeah, just trying to Decode this for you. All right, Buddy, we got 26 minutes left. Hopefully you can tell your crew that you're going to be late. Let's go into. I don't know if this is, this is probably my favorite topic because like – it pits Dan against Pat on The Flip where we come in and take polar opposite positions on very nuanced topics, even if we believe it or not. 

‍

Daniel Newman: Simulation. Simulation. 

‍

Patrick Moorhead: And this week we are going to be debating and talking about is Intel threatening not to do 14A foundry without an upfront CapEx commitment from a customer? Is this a brilliant move or a boneheaded view? Let's roll the dice. 

‍

Daniel Newman: Flip the coin Morehead, not roll the dice. It's a freaking coin. 

‍

Patrick Moorhead: I got you coin. 

‍

Daniel Newman: By the way. 

‍

Patrick Moorhead: The only competition I have, this guy, Snake Eyes. That's what I was thinking about, Dan. You and Snake Eyes and then the dice. All right, I'm going to be arguing that Intel threatening to not do 14A foundry without an upfront commitment is a brilliant move. Dan Lip Bu Tan, the ultimate negotiator, pulls a move very similar to what Qualcomm and TSM – sorry, Qualcomm and Broadcom did with Hawk Tan. So what did Hawk do with Apple? Apple was waffling on buying Bluetooth and WiFi. And then it came out that he said, well, if we can't do this, it's not a good business, we're just going to shut it down. Which, I don't know, a couple weeks later, a couple months later, we see an agreement that Broadcom and Apple have come to a long term agreement on WiFi and Bluetooth. Success. And then Qualcomm, right, with Apple selectively leaking out, right, it's doing its own modem chips, which it is. We're gonna get there, we're gonna get there in a year. Oh wait, I mean two years, I mean three, I mean four, I mean seven. Right. And Cristiano (Amon) called them on that and basically floated that said we may not be even doing a thin modem for Apple in the time frame. We all know that Apple needed the modem. And then shockingly, right, we had a licensing agreement and a modem agreement that was long term, so it totally worked. So this is essentially that. And what I want to peel apart is, is in reality, this should scare the heck out of NVIDIA, amd, Apple and Qualcomm and Broadcom and Marvell. Because if Intel doesn't do this, we will see TSMC prices and profits, profit on a percentage and profit on a dollar, absolutely skyrocket. And they're just not thinking about that being a reality. And I think they're looking for somebody else to do it. They want somebody else to go first because right now what they're looking at is what's right in front of their face, which is opportunity. The other thing is going to be innovation, where if Intel doesn't do 14A foundry, you have less competitors, less people doing it. It's pretty safe to say you're gonna have less innovation. I mean look at, you know, look at, look at markets where there might be just one player, maybe one and a half players out there. The geopolitic, the geopolitics are immense and the U.S. government doesn't have a clue about this. Trust me, I've talked to them and that is – foundries do two things. They create IP and they make chips based on that IP. All of the IP for TSMC foundries is done in Taiwan. All 10,000 of those engineers. C.C. (Wei) did that, elucidated that on a conference call. And that was also rung through the Taiwan version of our House and Senate. And therefore if Taiwan gets invaded and the U.S. gets cut off, we're going to be stuck on whatever the next manufacturing node is, probably without the capabilities to improve on the current nodes. And then TSMC is basically going to be like SMIC is right now. The inability to move stuff forward because of the lack of IP. So, I think we will start getting – I'll bet you Kevin, Kevin O. Buckley. His phone was ringing off the hook like, well, wait a second. Well wait a second, right? And then, you know, if I'm Kevin, I'll be like, “Hey, you've just been negotiating and using me as a lever to get a lower price with, with TSMC. So it's time to, you know, buck up or shut up.” And I think it's going to be a brilliant move. 

‍

Daniel Newman: Oh, well, I just did a check on my, you know, history and I don't think that the Chinese or the, the People's Republic of China or Communist China, whatever we want to call it, has any real claim to Korea. And I did just check. And Elon Musk just committed some of his most important foundry and fabrication to Samsung at 2-nanometer. And so while we're so busy claiming that Taiwan and an invasion would be the end of innovation, we forget that there's another country that is Korea that has a leading edge Fab and Foundry. And that, yes, has had a little bit of a struggle, but is clearly breaking through by gaining the trust of someone like Elon Musk to build scale on their leading edge. No. So yes, there is an argument that the U.S. needs to have its own chip champion. And heck, I may even know someone that's made that argument. But we have diversification. We have work being done with IBM and Rapidus in Japan that can also handle 2-nanometer fabrication. And as far as I know, China doesn't have any claims on the Netherlands. So high NA and EUV will still be available in other markets. And so sure, yes, Taiwan is the best right now and yes, there is some risk there, but there are other players and Intel making such a stake is basically just them putting a throwing a white flag out there saying, “Hey, U.S. Government, you better back us because we're threatening to leave the market if you don't give us some, give us some money, give us some support and maybe just force our, force the customers that don't actually want to work with us to work with us so that we can actually continue on this path.” But it's very simple. The options do expand beyond Taiwan. I'm sorry, Samsung's a more than capable company, and yes, Intel in its heyday certainly had the potential to be a Foundry. But after this 18A debacle, I doubt customers are confident, I doubt they're going to be investing heavily and I'm pretty sure that the U.S. isn't going to let China invade Taiwan anyway. So I don't know, wrong move for me. I think they needed to go do it the right way. Build the best node, make it high performance, give the customer, no reason to select anyone else. Put some pressure back on Taiwan to not be so greedy with their high prices. But if anyone else gets it right, and I think Samsung will, that will be enough to put Taiwan back in check. We will be fine with or without Intel. 

‍

Patrick Moorhead: That was good, Dan. And I actually did, I actually did cite Samsung and my original tweet that got I think 17,000 views on it. But yeah, yeah, this is a very nuanced, nuanced issue here folks. And the other thing I want to point out is that Intel is going to do 14A for its own silicon. There's a lot of extra expense that you need to lay out capacity, right? To even get the third, all the third party IP from let's say Synopsys and Cadence out there rolling. 

‍

Daniel Newman: Can I say something just for the record because I'm going to get probably a lot of Intel bulls that won't realize. And remember this was a simulation and I'm just doing my awesome debate stuff but like I want intel to be successful. Like it is very well documented that I would like to see Intel execute. I wanted to see them execute on 18A and I was very disappointed in how that's all gone down. But 14 was always, you said, you know, it had the mobile, it was more like there's something you'd been on the record many times talking about higher likelihood of being able to support mobile SoCs certainly gives some more time for very sensitive, high volume, custom type chips that, that are being done by tsmc. And again, none of these companies, they're all willing to pay the 25% premium to not mess up yield and not mess up performance. So these companies can't afford to get wrong right now. So that was why it was so interesting with Tesla. But if anyone has that kind of latitude to, to, to accept a lower yield. And I absolutely have to imagine, Pat, that Tesla got one heck of a deal, one heck of a commitment for support for volume. And of course as Samsung fixes their problems, Tesla is going to get the volume it needs in the future for basically taking this risk because no one else at this point has been willing to take that risk at any sort of scale. 

‍

Patrick Moorhead: Yeah, what I'm, what I'm hearing in the ether is that it was a combo ASIC D-RAM and a D-RAM deal. So all those robots, all those cars need, need more memory too. 

‍

Daniel Newman: It's solid not as some of the other ones, but it is still on a 2, which is not something we've heard a lot about from Samsung. 

‍

Patrick Moorhead: That's right. All right, good discussion. I know you all know I win these things all the time, but, Daniel. 

‍

Daniel Newman: Do you think you won that one? 

‍

Patrick Moorhead: Yeah, you did a really good job there. 

‍

Daniel Newman: That was fun, though. 

‍

Patrick Moorhead: Yeah. All right. All right, folks, let's go into the Bulls and Bears. We're going to talk about a bunch of earnings, a ton of tech earnings last week. And also tariffs came up again, and nobody talked about them until yesterday. We were all, you know, jumping around, doing high fives. And of course, the Powell interest rate saga continues. So let's dive in. All right. Hey, in my Week Ahead video that comes out on Mondays, I'm like, this week's gonna be all around about earnings and tech earnings and tariffs. But, Dan, really, nobody was. We were doing, like, happy dances around tech earnings. And then tariffs hit last night at about 10:30. What the heck is going on here? 

‍

Daniel Newman: Yeah, I mean, look, I kind of hit it at the top because I just felt it was more important than whatever that first nonsense topic we had. But, you know, Trump kind of did what he said. He just came up with a list and said this is what tariffs are going to be. And I think, you know, there's been this TACO trade that Trump's going to wait and make deals that everyone's happy about because no one, it's like they don't care about the rate as long as they feel it's bilateral. But when Trump starts kind of talking about just applying a tariff to a nation, that's when people get all sensitive. I still think this is a bit of an algorithmic dump. I think August is always a bit of a down period hit right at the right moment. But Pat, like, I think overall the trade war is going very well. Everything is pointing to there might be some inflation in pockets, but it wasn't. Like there weren't inflation before. But at least we're raising revenues and it's not through tax. Like, I don't know about you, but, like, I'd rather be taxed a little bit more with inflation on what I consume than tax more on money I've made and haven't even had a chance to spend yet. 

‍

Patrick Moorhead: Yeah, I agree. It's, you know, I don't even know if the market is down based on jobs or is it down actually based on tariffs. When I look at tech tariffs, I did a little research on this this morning, and it looks like right now. So first of all, almost all U.S. AI servers are done in Mexico, and that's already covered by USMCA, zero tax chips displays are still sandboxed, right? Haven't been decided. But there's the section what, 232 that hasn't been decided on yet. That that could impact in the future. And plus, you know, China got pushed, got pushed down, kicked, that can got kicked down to August 12th. But for right now, China and Taiwan, converted notebooks and converted phones – our sandbox. So ironically, right, it's Apple and smartphones that are done in India that are at 25%. I've got two or three sources that, that says that it is. And from India, nothing particular from the United States yet. But that would be pretty devastating to, to Apple. So let's talk about interest rates here. Powell once again got up and essentially said, you know, too much uncertainty about what's going on. Inflation is not where we want it, it's at 4, we want it in, in the 3-ish zone. And then you saw some interesting folks from the Fed come in and have some conversations about saying it without saying it, that we should reduce interest rates. And you know, Dan, that does not happen a lot out there. And then we've got this Jobs number that hits, that was off by a freaking mile. What was it? It was supposed to be a net-add of 147,000 and I think it came in at 14,000. How the hell can you be off On a factor of 10? Like how? No idea. 

‍

Daniel Newman: The adjustments on most of the data is pretty palpable. So anyone that pays attention, we read out the numbers, everyone reacts to that and the market tends to not react very much to adjustments. And so, if you actually look at Jobs data over a period of time, the adjustments have been revisions down pretty consistently. So, you know, when they're pumping the economy, they kind of put out data with a margin of error and then they quietly revise the numbers. It's a, it's a great tactic for, you know, winning “Popular” because most people don't actually pay attention to the revised number. Pat. Like, look, you know, we've heard about it, we felt it. And the economy cuts, reductions, AI is impacting jobs right now for sure. High interest rates, lack of, of, you know, but, but again, overall the economy's been better. So inflation is down, growth has been better, GDP was up. So it's kind of like, I just think based on how expensive assets are and the fact that our, our macro strategy is to outgrow inflation, then we need to bring interest down to one, make debt more serviceable and two, stimulate spending because like people can buy more house, people can buy more cars, people can spend like so if we want to stimulate spending like high rates are a limiter. People don't move people especially like housing and stuff because like so many people are in these low locked in mortgages from when that we were close to 2% or 3%. Like you want to stimulate people to move, you gotta, you gotta make the rate competitive because nobody wants to get half the house for twice the price. It's like, you know, that's why no one goes to California anymore. Right? Half the house anyway, that's it. 

‍

Patrick Moorhead: Lot a lot of tech earnings that we need to grind through here. Let's start off with Amazon. They just crushed it, absolutely crushed it on the quarter and they had what was considered a soft guide. There was also some discussion on the conference call that made investors a little bit weary, kind of a doom and gloom in the way that some of the questions were, were answered here. And you know my top-line here is that that this company is absolutely a, a rocket ship company that is poised for growth. Kuiper, which is the competitor of Starlink is not even baked in at all. Zero value. The fact, I mean I, I nailed what the growth was going to be on AWS. It was at 17.5% which on $125 billion is an absolute monster number. You know Azure's at $75 billion and that includes OpenAI. Nothing to take away from Microsoft here. But I just have to, I have to, I have to point that out and from a tariff perspective they, they on Amazon.com, they seem to have pretty much gotten completely around that. Look at the North American numbers. They're off the, absolutely off the charts. And we can debate whether those were pre-tariff, pre-tariff buys. But you know, I think like what do you say Dan on you know, Halloween masks and, and Christmas trees and you know all the, all the stuff that shows up on my doorstep every day. They seem to have absolutely weathered, weathered it so far. Final comment. Is advertising 23% growth on a multi-billion dollar business. Third largest digital ad platform on the planet next to Meta and Google. 

‍

Daniel Newman: Yeah, I mean you hit most of it. The backlog's a really interesting debate because their growth is so much slower than Azure and Google. And so there is, I did the calculation four years from now at this current growth run-rate, Microsoft will be bigger than AWS if they keep this steady. So it's, it's the, the gap is narrowing. Azure put out their number for the first time but the backlog is really big and the constraint was they said largely energy. But I, but I also have known from the onset, AWS did get a little behind in terms of volume of NVIDIA. Some of that was based on its own chip and I think they had a little bit of that sort of principle office moment in terms of getting the early capacity but, but energy constrained and it seems that you know, somehow Microsoft has done a better job but that bigger backlog could start to fill a gap could fill and their growth rate should accelerate. 

‍

Patrick Moorhead: Alright, let's move to a company I don't care about at all. 

‍

Daniel Newman: I, I, I, I see this one in here and this is just me talking my book so I'll keep it really, really brief. You know my biggest personal position is SoFi. This company is, is revolutionizing Fintech. It's one of the fastest hyper growth and by the way it's also like tripled in the price in the last couple of months. So really, really glad I'm in that position. But they absolutely just hammered their numbers on every, every side – loans, Fintech, growth. And of course then right after they had this incredible beat and huge jump they offered a one and a half billion dollar shelf that night and that's a, basically they diluted themselves so everybody went crazy. Stock went down like 7% immediately but then it got filled in. So it's clearly a lot of momentum here. I won't say much else because it's a little bit off the, the beaten path of our, of our audience but you know I'm not making advice recommendations but I would buy this stock. But I'm not recommending it. But I would buy it but I'm not recommending it. 

‍

Patrick Moorhead: Yeah, hey, let's move to Apple. It's fine. 

‍

Daniel Newman: You don't care about them, do you? 

‍

Patrick Moorhead: I do. You know they're Mag 7. You know one of the points I've been making on broadcast is you know, are they, should they even be part of the Mag 7 anymore? From a market cap, yes. From a growth and excitement from AI, absolutely not. They had a very good quarter particularly around iPhone. Management said about 1% of that was from the “buy ahead,” and I think if you look at the overall kind of 10% growth that that wasn't a lot. Macs really crushed it at about 15%. They had a new lower priced Mac that that came out and they got a lot of uplift from, got a lot of uplift from that. Second thing, is they had 4% China growth and everybody's like “Oh my gosh they figured out China.” Here's what happened. Apple was included in the government programs that subsidized Apple for the, the first time. This is why it did well. It's because relations with China are thawing and China's like, okay, we'll put them on the, the list as, as well. So they're down. And why are they down? For the reason that I listed in my tweet this morning, which is, I know the entire market's down, but if you have your best iPhone quarter in how many years? Like four years? Shouldn't it be up like 10% - 15%? Right? And then you've got the tariff piece. They're getting hammered. I believe that people are finally realizing that the tariffs are going to clobber, are going to clobber them. 

‍

Daniel Newman: Yeah, I'm so, you know, Apple's, uh. It’s by the way, it's set for a pop because like the minute they actually do something in AI, they will get a huge bump. But everything about it was just a value company. We're just doing good with our inundated embedded products. But the big thing was Tim Cook did open up the door about a possible being accretive and doing deals and leaning into AI. So he did say kind of what I think he needed to say, but now he's got to do some “Prove it.” He's got to do some prove it. Or I like to say “Over to you, Tim. You're gonna have to figure this one out.” But it was a good quarter for them. I mean, every Mag 7 company, except for Tesla, who doesn't? Nobody cares about fundamentals anyway. Did the business. 

‍

Patrick Moorhead: Yeah, totally. Are you shocked at all that they're down right now? 

‍

Daniel Newman: No, I mean, most stuff's down right now. 

‍

Patrick Moorhead: It is. But you know, I look at the comparison on the percentages of, of, of stuff. They had the best iPhone quarter than they've had in years. 

‍

Daniel Newman: Yeah, well, you know, R.J. Reynolds sold a lot of cigarettes last month. You know, that's the comparable. I saw a lot of Jack Daniels. I mean, there's nothing in that. That's just. It's great that their core business is there, but the valuation already rewards that. 

‍

Patrick Moorhead: Yeah, let's go to Microsoft. Pretty much a flawless execution here. I mean, not only did they – they had multiple beats, but their guide was huge. Satya took his $80 billion in capex and raised it to $120 billion. That was super, super impressive. For the first time, we got a peephole into the actual number for Azure, which was $75 billion at a 34% rate a little less than twice the rate of AWS $100 billion in bookings right that's up 37% Microsoft invested early they vested, invested ahead, and now I'm kind of scratching my head Daniel, on did they tell OpenAI to go find other cloud providers like Oracle and Coreweave or to to make room for their own business or is this something else? And when I kind of drilled in there there is some real enterprise demand AI, and surprisingly to me non-AI, you know we saw some huge numbers in in Copilot not sure not as not as big as Chat GPT but remember Copilot is really about it's really. 

about commercial.

Daniel Newman: Yeah this was a all about Azure you know that was the that was why it popped $70 - $75 billion 39% growth out clipped every other of the big three by a lot and like I said executing without infra of their own and with the OpenAI partnership still in sort of a weird flux but again the stickiest moat and enterprise is Microsoft and they're proven there were so many very impressive overall quarter and the thesis for the bull market and the AI stability has not been harmed by anything that happened in this quarter. 

‍

Patrick Moorhead: Yeah Meta absolutely astounding, and if there's any company that that that puts the exclamation point on downstream impacts positive impacts from AI expenditures it is Meta. They are the clear leader in this where what they're doing is driving stickiness in the platform and and adds up gigantic they increased CapEx even more they talked about 5 gigawatt hyperion data centers just an absolute machine and they we still haven't even seen the impacts of them money whipping all these researchers out there.

‍

Daniel Newman: It's the New York Yankees of AI they're buying the they're buying the trophy but there's a trillion he's spending a few billion to create a few trillion. It actually makes a lot of sense. It's not for sure and again the Yankees didn't always win no matter how much they spent but that's what's going on there but yeah best application of AI in the business. I was asked on Bloomberg to pick between Microsoft and Meta after and it was a hard one especially after Microsoft delivered but I still think Meta is the king of AI right now. 

‍

Patrick Moorhead: Some highlights for Arm so first and foremost they had a really good quarter the guide is what got people but the thing you need to focus on in the guide it is invest ahead for CSS. Meaning they're getting more designs that we're going to yield a lot more high profit dollars down the line but they're just, they're, they're investing more. The number one interesting thing though that came out for me were without saying it, Rene Haas essentially said they're getting into the chip business. He didn't say it but he came you know down to the one yard line, not in the end zone, one yard line from saying that he's going to do that. And net-net if successful, that would drive revenue. Huge revenue selling chips versus IP is a lot higher. Percentage on the margins are lower but the margin dollars are significantly higher. And they've gotta wade through selling their own chips if they do this with giving room for their current IP customers to differentiate. 

‍

Daniel Newman: I think the company is so embedded and so sticky it has some latitude and yeah, could this push some people to risk 5? Sure it might happen but Arm is still the dependable IP provider for so many critical chips. And I think the pressure for CEO Rene Haas is to grow it faster, grow margins faster because I mean remember this thing trades at like a very high valuation $4 billion in revenue and it's trading at like $150 plus billion in in value, and he needs to grow faster and AI is you know he's adjacent to AI like CPUs for AI but like getting directly in and playing that game is, is a big market opportunity and I think they think they have enough asymmetric power over the, the current licensee and those that use its technology and I think it also feels it can fill a gap that exists there and being the partner to help develop chips. So I think it's the right one. 

‍

Patrick Moorhead: Dan, they have 50% share now in data center without latest NVIDIA that's all Arm in there. 

‍

Daniel Newman: Yeah, that's why I said like they think they can pull this lever if they want. You know it's like we dare you to go somewhere else. We'll see. 

‍

Patrick Moorhead: Exactly. And we really haven't seen any of the Stargate yet. But anyways final topic. Qualcomm, Dan, how did they do, how did the market react? 

‍

Daniel Newman: I mean the market just won't give Qualcomm a break. Quarter after quarter beating 15% growth outside of without the Apple number in 10% or so with the that number we all knew that was kind of moving away strength in you know the IoT the Automotive portfolio. So overall like I said a good, good you know performance by the company. This is one of those like where and we know Cristiano well respect him very much. Like I think this is like a showman moment for this company though. Like everybody's. It's a little bit like Apple and handsets. Like people are just like, what is next? Where does. You know they're in this autonomy. You know, I keep saying Autonomy for Vehicles, XR, IoT and Physical AI SoCs for Mobile. I just keep thinking like they need to be like the robot company. Like they need to be the future of robots and Autonomy sort of change the narrative what the company leads with because it just trades. It's such a, you know, it's treated like, like an old value, antiquated licensing and IP and handset company and people are waiting for something more. So I think there's like this kind of showman, this musk kind of moment where Cristiano needs to come out and tell a story that's like five years into the future and how, you know, maybe Optimus will be the $30,000 robot with all the technology that sits inside of the stack at Qualcomm. The. The billions of $5,000 robots that are going to be sold in China and everywhere else are going to be powered by Qualcomm. Model that out analysts. That was my bet. That's what I said on Bloomberg. 

‍

Patrick Moorhead: Market wants data center infrastructure or AI growth. Now Qualcomm's in a tough spot. They discussed data center in Middle East and then Cristiano had some words about revenue when revenue will hit Fiscal 28 - 27 calendar. We are seeing AI lift in smartphones evidenced through ASPs and premium growth. Growth. It's not huge, it's not a super cycle. And the market doesn't see that currently as interesting. I think the market is. You know, I agree, Dan. They are missing that big picture on the Robotic Edge. And what Qualcomm needs to do is get a robot on stage, announce some design wins with companies that people view will be winning into the future. You know, we did see some evidence of this with Amazon and some of the robots. If, if you remember, at their last event, one of the walk-on videos was Amazon talking about a robotics edge. And then we saw Amazon that said, we've got a million robots out there, we're the biggest company on the planet with the most robots, but that has to be put on there. Finally, they need to give details on their data center products, otherwise people aren't going to believe anything.

Daniel Newman: That's a good call out. I didn't talk about that, but yeah. 

‍

Patrick Moorhead: So folks, thank you for this. We're on. Gosh, a minute minute, an hour and eight minutes. Appreciate you hanging in there. Hit that subscribe button. Have a great week. We appreciate you. Hit Dan or I up on X. We spend a lot of time on there and that's where to get our quick takes. Take care. 

‍