On this episode of The Six Five Webcast, hosts Patrick Moorhead and Daniel Newman discuss the tech news stories that made headlines this week. The handpicked topics for this week are:
- Dell Technologies Analyst Event 2024
- Qualcomm Q4FY24 Earnings
- Arm Q2FY25 Earnings
- Coherent Q1FY25 Earnings
- Lattice Q3FY24 Earnings
- Meta Llama For Defense
- Trump Expectations on Tech
- Solidigm 122TB SSD
- Cisco Q1FY25 Earnings
For a deeper dive into each topic, please click on the links above. Be sure to subscribe to The Six Five Webcast so you never miss an episode.
Watch the episode here:
Listen to the episode on your favorite streaming platform:
Disclaimer: The Six Five Webcast 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.
Transcript:
Patrick Moorhead: The Six Five is live and we are back. I know you missed us. I hope you missed us, but a lot of traveling going on. Dan, you are out of the country, now you’re back. I was traveling. I mean, just a bunch of stuff going on. Dan, are you in an undisclosed bunker somewhere?
Daniel Newman: I’m in an undisclosed bunker and unless you follow my Twitter, you would have no idea where I am. But, yeah, I actually was just at CNBC taping something. I did my all-hands in the studio, because I had an all-hands meeting and I was running late, so I stayed and did it in the studio. So that was pretty fun. And then I sat in traffic. But, yeah, buddy, I literally have not been not on an airplane for five days in a row now. I’ve been somewhere. It’s nuts.
Patrick Moorhead: Well, hey, I was taking advantage of a middle seat for four hours yesterday.
Daniel Newman: You did good.
Patrick Moorhead: Yeah-
Daniel Newman: I was glad you shared that experience because I didn’t know what your lap looked like.
Patrick Moorhead: I know. It was a very humbling experience to not get upgraded to first, I couldn’t buy my way into first. I felt like bringing out a lot of cash, offer somebody their first… Buy their first class seat from them.
Daniel Newman: You ever seen Home Alone? Remember when she’s trying to get home, she’s like, “Take my earrings, my watch, take it-
Patrick Moorhead: Yes, yes. I was not giving up my watch. I was actually-
Daniel Newman: Can I give you a sob story? Can I give you a sob story?
Patrick Moorhead: Yes, please.
Daniel Newman: The WiFi was down on my flight to New York.
Patrick Moorhead: How did you survive?
Daniel Newman: You have to understand that you don’t really realize the crack and how bad this really is until you can’t get it. Well, first of all, I was trying to use satellite, because that’s how desperate I got, because there was a couple of texts I really needed to get out. But I did survive. I watched four episodes of Narcos: Mexico because that was the only thing I downloaded on Netflix-
Patrick Moorhead: Yeah.
Daniel Newman: But I’m telling you, I really do sometimes think about like, “Wouldn’t it be great if I could just… Maybe you do because you actually have a pretty good habit health-wise of turning off your notifications. And I don’t know if you still doom scroll, I’m not there if you’re actually doing that, but I always say it would be great to just watch a show and stay focused, because I have this ADD thing where I’m watching and… Bottom line is I could not sit there. It felt so slow to me to not be doing two or three things at one time. And that’s a good show. That’s a pretty fast-paced show too.
Patrick Moorhead: Totally. Hey, let’s dive in. We are the Six Nine, nine topics. So it’s pretty crazy here, but we have to pack this stuff in. We’re talking Dell technology analyst event that we attended. Qualcomm, Arm, Coherent, Lattice, Cisco earnings. We’re going to be talking about Meta Llamas for the defense industry-
Daniel Newman: Are you yawning?
Patrick Moorhead: I did. You caught me there. Just getting that out there, I-
Daniel Newman: Usually, it’s when I’m talking though that you get bored, not when you’re talking.
Patrick Moorhead: No, I’m the most interesting person I know, just ask me.
Daniel Newman: Yeah, absolutely.
Patrick Moorhead: We’re going to be talking the market’s expectations, Trump on tech, you and I both got hit up that big time out there from reporters. Solidigm introduced a gigantic 122-terabyte SSD. The Six Five was there and our video will be coming out soon. We were on stage, Melody Brue and I. So, Dan, let’s go. We are diving in. Dell Tech analyst event here in Austin, Texas. My favorite destination.
Daniel Newman: Yeah. First of all, Pat, it was great to be at home in a way. And if I wasn’t leaving for London the second day, it would’ve been a really great event, to be able to go home every night, sleep in my own bed. This event was under NDA, so we’ll have to keep it a little bit high level here, but we are at the precipice of a megatrend right now, some things that are very specific to what got announced at the event. And, of course, Dell has this really interesting tailwind right now with a certain company named Supermicro that can’t seem to file its reports right now, but we spent… It was a very NDA event. It was three days. It was tons of access. You can check our Twitter, you’ll see some photos. We had Michael Dell there, we had Jeff Clarke there. We had the entire ELT there. And they were really giving us their vision of how they’re going to stay competitive, how they are innovating for the AI era, how they are investing in software and new innovation, new competitive posturing. It was a really, really well-run event.
And, Pat, something I really appreciate about Dell is that they really do such a great job of making their executives accessible, giving us adequate… Ample time, in fact, and just really figuring out how to keep the cadence down to amounts of time. I don’t know about you, but 45 to one-hour sessions diving deep into things, whether it was generative AI, whether it was sustainability, whether it was new data, infrastructure, architecture, and then, of course, the executive sessions, the Q&As. And, of course, the time on the lawn at the… Was it the Omni? It was a great place out on the back. And what? You and I got a couple hours pretty much with Michael and Jeff to talk about all things. And this is the stuff we really wish we could tell you, but can’t, sorry. But, Pat, yeah, I mean, this is a bit of a flyby, but it was a really, really overall positive event.
I actually got to go on a tour. I went on campus at Dell, and this is not under NDA, but I got to see some of the stuff they’re doing with direct liquid cooling, super cool stuff. The data center’s changing so much right now. I mean, it’s changing so much so fast and Dell seems to be really capitalizing on it. Of course, there’s complexities and there’s new competition, there’s a lot of silicon diversity, there’s a lot of innovation and development that’s going to be going on. A lot of competition with the cloud that the company’s going to have to face. But I think overall, really positive event, and I’m going to keep this one pretty short because, well, it was mostly under NDA.
Patrick Moorhead: Yeah. Good commentary. One thing I really appreciated about this event, aside from it being in Austin, was… What are you taking a picture of, dude?
Daniel Newman: You. I was taking a picture to tweet that we were podcasting.
Patrick Moorhead: That’s nice.
Daniel Newman: Yeah. You got excited.
Patrick Moorhead: I really liked the senior executives walking around, the presidents of all three divisions. Michael and Jeff Clarke were there. And that means a lot to me. When you hide your executives, it’s that you’re probably hiding something and they don’t want hard questions. And if executives take hard questions, that gives me confidence, they can deliver the goods to the end customers. And then there’s strategy, right? There’s execution, but there’s strategy. And comparing and contrasting to what I saw last year, I am a lot more impressed than I was the year before. And I do recognize that enterprise AI is changing a lot and we’re all still trying to figure it out. And quite frankly, it won’t scale probably for a couple years. That doesn’t mean it’s not… The building blocks aren’t going in there. But when we see mass adoption, it’s going to be probably a couple years off. And I do believe that Dell gets enterprise AI at the subatomic, meaning the piece parts, and also at the 10,000-foot level as they roll out all of these new innovations in the future. I think you will probably most likely agree with me that what Dell is bringing out surprised you. I know that I was surprised.
Daniel Newman: I wish we could spill, but unfortunately, y’all are going to have to wait a little bit. But there is a lot going on. It’s going to be a worthwhile conversation when we can come out with some of these things.
Patrick Moorhead: Yeah. Let’s move to the next topic, earnings. I want to say just upfront that we are talking about publicly-traded companies. Don’t take anything that we are talking about and use it as investment advice. Find a professional. So Qualcomm, I mean, had a great quarter on the backs of really all three businesses. I think the standout was auto segment with that gigantic growth and delivering on that $45 billion backlog. IoT was an absolute factor. A lot of product announcements across fixed wireless access, PC and XR. I’m really excited to see what the company will do in the future with AI on the industrial edge. And people aren’t talking about this, I hope to hear more of this. Gosh, next week, at their annual Investor Day, I’d remiss if I didn’t bring up smartphones where their growth is probably 2X the market growth. I think this bodes well for Snapdragon 8 Elite smartphones that we saw, does look like a financial winner. Now all we’ve seen so far is Chinese handset companies, but my expectation is that we’re going to see Samsung, which I think had to do with the positive guide and forecast.
Daniel Newman: Yeah, Pat. It was really positive overall. We know the handset business, the PC business, it’s holding up. And for Qualcomm, that’s the really important thing right now, and having that diversification. You and I have talked about over the last couple of years how important their move to diversification has been. And, really, the quarter got carried through by really strong results from the auto business, which you pointed out. I mean, [inaudible 00:10:30] basically design wins from every major automotive company. And now what we’re starting to see is that $50 billion design pipeline that we’ve been touting for the last year turning into quarterly revenue, the acceleration. And every quarter, that revenue is picking up, picking up. And [inaudible 00:10:44] out there doesn’t know about this, right? It’s not like you get a design and you start to produce. It’s a design win, and then there’s years oftentimes between the design win and when that design goes into place.
So the other thing is all these cars have lots of content being added. So every year, they’re adding more content. There’s hundreds, sometimes there’s thousand chips in a vehicle and people don’t always realize that, but that creates a lot of opportunity. And then on the other part is the industrial, the IoT business, sensor business. That’s come back. So those two things really carried the day for the company. And then, of course, the core businesses were intact, so they were able to beat the expectations, beat the guide, and the market responded, I thought, really positively overall. I think it was up 9 or 10%, the day of earnings. And we apologize. We know this was last week’s news, but come on, this stuff doesn’t get old that quickly, does it? So, Pat, I’d say, overall, the diversification strategy is working. The company’s looking like it’s in positive shape and probably the biggest question mark around the company is what we’re going to talk about next.
Patrick Moorhead: That’s right. Let’s dive into Arm’s earning.
Daniel Newman: Hey, did you like that transition though? That was pretty-
Patrick Moorhead: I did. I actually was paying attention and not looking at myself in the video, which is hard to do.
Daniel Newman: You do that a lot? I actually minimize myself because I know I’m just not that handsome, but I’m going to get with this new trainer because he’ll make me buff. By the way, if I work with Martin, if I work with him and tell him… Maybe you can show them this clip. I’m going to only have them do biceps. I’m going to be that guy with two-
Patrick Moorhead: Yeah.
Daniel Newman: … legs and just giant biceps. All right. So let’s talk about Arm a little bit. The point of contention is this ongoing dispute. We’ve weighed in on it in the past, not going to spend a lot of time here. Let’s just say it’s a ongoing question mark that’s going to need to be addressed. But the company is seeing acceleration in revenue based on strength in AI. It wasn’t a huge jump, but it was a beat. There was a very modest raise on the annual guide. The market was mixed in terms of how it responded to the results. But let me just share what at least my take is on why Arm is going to do so well. One is Arm is one of those companies that no matter what the part of this AI narrative and wave that you’re looking at and thinking about, they’re a beneficiary.
If you’re thinking about it on handsets, they’re in every handset. If you’re thinking about it on PCs, obviously they’ve got a very significant role to play in PCs. And again, not just the big SoCs, but ARM has lots of other parts and pieces too that go into these devices. If you’re in automotive, they have a role in automotive. If you’re looking at all these sensor devices, the IoT, they’re in sensors. If you’re looking at servers, they’re in servers. They’ve got lots of different parts and pieces for AI servers. They’re not the main GPU, but the head nodes on these servers. Of course, the core data center compute, the data center networking. They’ve got lots of parts and pieces. So every one of these trend lines where we’re seeing more silicon content, Arm has a role to play and Arm is making more money in a couple of different ways. They’re making money with royalties and they’re making money on licensing.
And it’s interesting because they’ve been spending a lot of their attention on subsystems. And as they get these bigger subsystems and as they start to provide not just licensing but also royalties to all these companies where they’re providing more important design content for more complex designs, they’re basically becoming an enabler for companies that don’t have all the in-house design prowess. And for those companies that want to… The Microsofts and others that are jumping deeply into silicon that want to do more, they’re going to use Arm, they’re going to pay Arm more royalties and they’re going to have more revenue streams, so the per device or per new part Arm is making more money. And this is really important too because a lot of people have been questioning, can Arm make enough profits. It’s a large revenue company, but really you’re talking about four-billion-ish on run rate a year and they’re creating about half a billion dollars in free cash flow.
So this is not a giant yet, but it’s a company that makes money on every device piece, part, component that you’re using and they’re doing really well. The pivot to v9 is going well, the… And by the way, just as an overall number that’s [inaudible 00:15:05] this in a little graphic, Pat, 300 billion Arm-based chips have shipped so far. So the momentum into PCs, the continued winning in handsets, the IoT devices, automotive devices, complementary server parts, the company’s in a good position. Of course, people want to see more growth tied to AI, but they are really involved in AI everywhere.
Patrick Moorhead: Yeah, great quarter. I mean, the expectations of the company are daunting. The company is delivering. It’s just that pesky guide that gets in the way. There’s so many ways that Arm can grow. I mean, even the markets they’re in, they can grow, in smartphones. Right? They grow with v9, the bigger cores, more bigger cores. With CSS, right? That we saw with MediaTek. And what’s interesting, I was very surprised to see MediaTek called out because you’re like, “Well, wait a second. MediaTek’s been doing smartphone SoCs forever. How could they benefit?” And really what this is, it’s about time to market. And there are other things, there’s more products that MediaTek can bring out by leveraging Arm’s CSS. And so you have CSS in smartphones, you have CSS in auto, you have CSS in server. Will we see APC CSS? Likely, that would be super-
Daniel Newman: Opens the door for entrance, and pace, and speed, right?
Patrick Moorhead: It does. And, obviously, the other way to grow is just when Arm grows market share in certain categories, it’s still… Even though it has gained a tremendous amount of market share in general-purpose data center CPUs, it is predominantly an X86 world. Okay? And I think the final comment is around v9. People think that when it gets to a large v9 footprint, there’s no more growth potential. But prices aren’t fixed and new designs come out and have a different royalty rate. So don’t think that even when Arm gets to 75% on v9, the growth doesn’t continue.
Daniel Newman: Good call. Yeah, I think that’s probably one of the highlights that I know Rene, because I think we probably both spoke to him around earnings, really wanted to see was that v9 is not the end. There’s lots of other ways the company’s growing.
Patrick Moorhead: That’s right.
Daniel Newman: Double-click, double-click, double-click.
Patrick Moorhead: Yeah, I appreciate that. Hey, let’s dive into Coherent earnings. And I think we gave a little bit of an introduction last time. I mean, we got interested in this company for two reasons. First of all, any company that’s engaged in the AI trade and driving AI forward is interesting to us. But also, Jim Anderson. Right? Jim goes to stuff with growth. He went from AMD when it was on a rocket ship, he went into Lattice, created that into a rocket ship, and here we are at Coherent. And by the way, if you haven’t checked the stock, it’s a rocket ship at this point. So I want to talk about coherent in context of AI. Right? Essentially, what they do is the lasers and laser solutions that are related to silicon photonics that transports a ton of that data over networks, both short haul and long haul… Long haul is between data centers, short haul is inside of the data center.
NVIDIA is a big customer, big hyperscalers are customers, and they don’t just sell the chips, they sell the entire enclosure and literally down to a chassis that has hundreds of silicon photonics links in it. And they saw 89% growth for 400G and 800G transceivers. And it wasn’t just revenue. They saw gross margin enhancements there. Industrial was down. And in the call, Jim talked about exploring divestitures of non-strategic assets to focus on high growth rates like Datacom. By the way, I think you and I have had this conversation, I don’t think communications is the way to talk about this. This is high speed data networking in case anybody was wondering in that. And I think there’s a Capital Markets Day for them coming up, and I’m looking forward to Jim and new leadership team. Sherri Luther is a new CFO. She came from Lattice. I’m interested to see what the leadership team brings out.
Daniel Newman: Yeah. You start to wonder about that domino, Pat. But, yeah, it’s good coverage there. I mean, my big takeaway is big-player silicon photonics. I think you and I are both trying to coach along the criticality of… Criticality? Is that a word? Criticality of terming data communications and networking as what it is, but the moving data at the speed of light for AI is a huge opportunity. This company makes modules that others use, and they obviously make them themselves directly and sell them. They’ve got some of the biggest companies as customers in the world. You look at the picks and shovels narrative, well, this is one of them that’s providing parts to companies. They won’t probably say it, but I’ll say companies like NVIDIA, companies like Broadcom are buying stuff from Coherent.
Under the new leadership, it seems they’re doing well. I like how they’ve gotten back to displaying their growth. We definitely got in on the growth 28% up, 300 bips of margin expansion. And by the way, this has been one where a lot of companies that aren’t core to AI just haven’t been able to do, is doing it. And by the way, increased EPS by 357%. So this isn’t necessarily Jim. I mean, he just got there. This was momentum. But he’s certainly getting back to making sure the right numbers are getting reported out there. So this company’s strong, Pat. I mean, it’s a really interesting one. It’s one that doesn’t get a lot of attention, but it probably should as one of those peripheral AI plays that has some significant upside.
Patrick Moorhead: Yeah. Good stuff, Dan.
Daniel Newman: Thank you.
Patrick Moorhead: Yeah. No, that’s good-
Daniel Newman: Back after each comment.
Patrick Moorhead: Let’s go into Lattice. Ford Tamer just came in as the new CEO for the earnings. What happened?
Daniel Newman: Yeah. So, I mean, he’s coming in at the opposite end, right? He’s coming in right now. I think everyone hoped for companies like Lattice that had high exposures to industrial to comms networks to consumer, and the people were hoping for V-shaped recoveries. The market’s jolting to all-time high. There’s all kinds of enthusiasm around AI. This turn has been a little bit more conservative. It’s happening in a bit more of this extended U-shape. What I do think is starting to become evident is that there is a bit of a bottom here. The company still is winning in terms of across the board in FPGA. If you look at other companies in this space, part of… AMD doesn’t really focus on this very much because of the diversification, the MI business, some of the PC growth in the business. But Xilinx is slow, Altera is slow, so Lattice is doing the best of a difficult situation right now.
So they’ve seen their margins drop, they’ve seen earnings… They’re, I guess on a sequential basis, doing okay, but on a year-on-year basis, they’re struggling a little bit. They’re doing well in parts, they’re doing well where there’s AI, they’re doing well in… Some of the stuff they’re doing around devices, the sidecar FPGAs on some of these new AI PCs are opportunistic, on phones, on servers. But overall, Pat, this is a company that is in the middle of a recovery. The good news is I do believe there is a recovery. But the key is also not just diversification, it’s that they’ve gone upstream. So they’ve built their TAM, they expanded their TAM, which is something they’re going to need to do and they’re going to need to continue to really double down and dig in to communicate to the market the AI play that they have on handsets, or on PCs, on server.
Because whether it’s Lattice, whether it’s Marvell, whether it’s Coherent, a lot of these companies are digging their way out by basically turning what used to be core businesses into secondary businesses and are creating these new businesses in silicon photonics, in networking, in AI chips. And so Lattice has the same challenge. The core legacy parts of the business may be a little slower. Automotive’s another interesting part of their business though too, which has been performing really, really well, but it is going to take some time here. This is not an overnight. They had 10, 11 great quarters in a row, and they’ve had now about three or four that… A little bit more challenging. In all the indications I got from the conversations I had was the challenges will continue for at least another quarter. But I think the diversification and some of the TAM expansion should serve them well, especially if markets turn in, if they can make that AI move.
Patrick Moorhead: Yeah. So what Ford outlined on the call, and by the way, I’m actually meeting with Ford tonight to say hi. I’d never met him face to face. I’m going to shake his hand.
Daniel Newman: Where at? Where at? Are you guys having dinner?
Patrick Moorhead: We’re just meeting for drinks right across from my condo. So he’s in town. But, no, listen, there were headwinds in industrial and automotive, right? A 7% sequential decline due to inventory adjustments and software demand. I know we talked about Qualcomm crushing it, but they’re actually not the norm there. Most of the industry, and this was some analysis that I had to do, which was, “Hey, we are seeing quarter-on-quarter growth from these industrial IoT spaces. It’s just there were softer demand and inventory adjustments.” Ford said he anticipates more of a U-shaped recovery in the middle of 2025 with modest single-digit revenue growth expected for the next year. There was a 14% reduction in force during Q3, not expected to impact any roadmap, things like customer support.
One thing I always like for a new CEO to do when they come in and give the lay of the land and what Ford talked about was financial discipline, aligning investments with high return opportunities, market leadership, strengthening the position in key sectors like communications and computing, you had mentioned that, and focus on edge AI. Capitalize on growing demand for AI. So, yeah, now it is interesting, if you look at Ford’s LinkedIn and his background, he is a hardcore M&A guy, right? And I don’t know what that portends to, what that means. Maybe it’s nothing. But he has a combination of being an operator but also being heavily engaged in M&A, and he has been a VC. So very, very interested to get to know him and see what he is going to do to drive growth and profitability at the company. Okay, let’s move to the next topic. So-
Daniel Newman: We’re ripping, dude.
Patrick Moorhead: We are ripping. So as you might know, Llama is really at the forefront in terms of popularity for open source models. Right? They have big models, they have small models, they have models for PCs, they have models for phones. And what we’ve seen a lot of companies do all the way from the big cloud companies to IBM to the on-prem folks is they are using these open source Llama models to be able to power their own customers’ AI. And one of the things that Meta did not allow was this to be used for any military or military-type contractors. But I wrote a piece on Forbes that showed, “Hey, they are now enabling that and aligning with some people that you might know. And one of the biggest reasons I think that the US government is getting around this is given the Chinese government $1.4 trillion master plan for AI that literally… And I’m not making this up, but from autonomous driving to automated factories to mass surveillance, and you can imagine that this would make it into the Chinese military complex. So anyways, in alignment with Accenture, AWS, Anduril, Booz Allen, Databricks, Deloitte, IBM, Leidos, Lockheed Martin, Microsoft, Oracle, Palantir, Scale AI, and Snowflake, they are all in on this. And what I liked about… Don’t yawn at me.
Daniel Newman: It’s been a long day, man.
Patrick Moorhead: No, I hear you. So what I really appreciated about this announcement, it just wasn’t like, “Hey, let’s throw the technology out there.” They outline Oracle talking about streamlined aircraft maintenance. And by the way, as you and I were sitting at the big Oracle world, Oracle’s first customer was the CIA. Scale AI, mission-specific enhancements. They’re all part of the Army’s robotic combat vehicle program. Lockheed Martin, I mean, come on, F-35, F-22, F-18, F-16. It’s crazy. Amazon Web Services, Azure, DoD Impact Level 4 and 5. It goes on and on. IBM Watsonx, US State Department, where they’re actually working with UNESCO for improved societal development. Meta has this thing called No Language Left Behind AI model that translates 200 different languages, even those ones in low-resourced and marginalized zone. So anyways, I feel safer because of this and hopefully you do too. But it also-
Daniel Newman: Are you serious or are you being funny?
Patrick Moorhead: No, I’m totally serious.
Daniel Newman: Okay.
Patrick Moorhead: I’m totally serious.
Daniel Newman: You’re a buyer. You’re into this.
Patrick Moorhead: Well, I am because, I mean, you know how long. You remember when Google wouldn’t do anything in the military and protesters? Now these tech companies are realizing that doing things that protect the United States of America is not a bad thing.
Daniel Newman: Yeah. Look, it’s good. I’ll give you a two side to this one. I don’t have as much depth on this. I didn’t write anything up on it, I just read it. I’ve got this Meta as a company that historically uses data in really interesting ways. You can argue about privacy and rights protections, how it gets… They’ve been an absolute leader in open sourcing the future of AI. So much credit to Meta with what they’ve done with Llama.
Patrick Moorhead: Yeah.
Daniel Newman: But I always think of companies that are good for defense as companies that are very trusted in their nature. Meta does not feel to me that way in their natural posture. They’re data for advertising. Having said that, I mean, I love diversification. I love this story. We all know what’s going on with DOGE. I had to make a DOGE joke, but Department of Government Efficiency, we’ll move into that here in a minute. But a lot of those contractors you named got clobbered in the market with Vivek and Elon looking to take trillions of spend out, and this is one of the areas. It’s these interesting forces all going on because… Like I said, I’ll hold my thoughts because we’re going to talk about Trump and what it’s going to mean here in a minute. But, look, I think Meta has been breakthrough for open sourcing, GenAI, making it available and making it so much possible. I just have an interesting reaction to the idea of them, data privacy and being involved in national security. So This is going to be an interesting one to see. But, I mean, look, the reaction of all those contractors meets a heck of a lot more than what I say, the ones that write the checks and do the implementation. So, I mean, it’s good progress, and like I said, in this case, Meta has been the… I think they’ve been the protagonist in at least generative AI and open sourcing and making this possible. So credit to them in that case.
Patrick Moorhead: Totally. Well, hey, let’s go in Trump version 2.0. You and I both got hit up… I mean, I got hit up on this when I wasn’t even supposed to be talking. I was supposed to be talking stocks and earnings and they threw out the new administration. So, Dan, what are your thoughts?
Daniel Newman: Yeah. I mean, this one could go long. I’m going to try to keep it short because I’m committed to finishing this podcast on time. But first thing I say is everybody that has an assumption of what the administration is going to mean is I think everyone needs to take a breath. I really do. I mean, I know we’re hearing about appointments, people are emotional, were charged about things. I want to talk more about tech and the tech and the economy because a lot of people… We’re talking about tariffs, we’re talking about trade issues, they’re talking about regulation and taking down big companies and Trump’s threats to TSMC, a lot of that stuff that’s going on. And my personal take on all of this is everyone needs to take a step back and think about what is Trump going to want this second term presidency to reflect when it’s over. What’s the legacy going to be?
And my take is the legacy is Trump wants to be pro-business, pro-growth, pro-stock market, and pro-technology. He doesn’t want to be the president that took down the US’s biggest tech companies through some regulatory teardown. I mean, of course, they’re going to have an SEC and an FTC that might potentially keep tabs on what companies are doing. He’s not going to want to be the one that took and slowed down the innovation around AI. So for instance, all the companies, the Mag Seven, you’ve got what’s going on in Taiwan with TSMC. You’ve got startups like Palantir in defense. A healthy market, by the way, S&P has double-digit percent weight from just seven companies, mid double-digits at this point. He’s going to want those companies to be successful, those companies to be growing. He’s going to want the stock market to be up double-digit percentages from the time he comes into office to the time he leaves office.
So when a lot of people are like, “He’s going to tariff everything and it’s going to crush this business,” I think a lot of it’s posturing. And I think, by the way, a lot of things he said… Everybody knows by now, if you believe that any candidate has been genuine in all of their campaign posturing, then it’s not partisanship anymore. You’ve just lost your damn mind. These people say things, they test things, they throw out ideas and they see what sticks and then they run with them like taxing unrealized gains, big tariffs on everything that gets shipped out of a country. I think he’s going to be attentive to US jobs, manufacturing jobs here. I think he’s going to push for more leading-edge semiconductor production here. Will he change the CHIPS Act? Well, maybe he’ll get the money flowing because honestly, it’s been a disaster in that capacity. Will he maybe put more dollars into Intel, and Micron, and US-based companies and maybe not as much into companies that are already nationally supported in their part of the world? I could see that happening.
But he knows that Apple needs TSMC and Qualcomm needs TSMC and Broadcom needs TSMC, and NVIDIA needs TSMC. I don’t see them throwing that out the window, but I do see a situation in which US-based companies get more incentive. I do see a situation where we try to create more symmetry with trade where these countries that pay nothing maybe have to pay something. I do see a very pro-lower taxes, business growth, market growth. So I’m not going to say I’m a Super Bowl, but I think Trump is a guy that wants to leave a legacy of a strong economy and a strong stock market as the indicator, therefore all his decisions will be made in the shadows or with those things in mind, Pat. So that’s what I think.
Patrick Moorhead: Yeah. I mean, Dan, that was great. And we don’t always agree on everything, but I agree on pretty much everything you said, but I describe it slightly different in that with Trump, everything’s a negotiation. Okay? He is not going to be the guy that wants to tank the tech stock market. And if you look at tech as a percentage of the overall market, it’s absolutely obscene and also knows that that’s into 401(k)s and regular people’s savings accounts. So do not see that happening. I mean, you want to put a 40% tariff on an iPhone? Okay, well guess what? That $3 trillion valuation plummets. Right? And there’s no pragmatic solution for that. There really isn’t. Apple requires a million workers, so many workers that China imports workers from Indonesia to live in China. Okay? It’s not going to happen in Mexico. I mean, Apple’s trying to move some of its manufacturing to Vietnam. Right?
Daniel Newman: India.
Patrick Moorhead: Yeah, India. It is a slow road. But notice that even China doesn’t have enough labor-
Daniel Newman: I’m glad you did say that though. I do think that is going to be the biggest antagonist in all… He is going to be tough on China. I have no doubt on that. And that probably will be the biggest challenge for all the companies that have double-digit percentages, Pat, in sales going into China.
Patrick Moorhead: Yeah, I think when it comes to TSMC, he’s going to be saying, “Listen, we spend this much money a year protecting your island, and I don’t understand why we would give you a cent to do that. Maybe we just shift that money somewhere else.” But in the end, it’s incentivizing TSMC to build more chips onshore. When it comes to even Samsung, I think it’s the same equation. And I do know there was some discussion, it was the speaker of the house who made this idiotic statement about getting rid of the CHIPS Act, which just was a huge disappointment. I don’t think this is a Republican thing. I actually worked on the initial research that went into the CHIPS Act, what became known as the CHIPS Act back when Trump… During his first term. And Trump was unable to get that across the line. Credit to the Biden administration for, not creating this, but getting it across the line.
Now where’s the check? These checks have not been sent. It’s an embarrassment. We should be embarrassed. And a lot of people think is that it’s a giveaway. Guys, look at the case studies from Upper New York where the state of government gave special tax treatment to companies like IBM, and now they’re giving it to Micron. They gave it to GlobalFoundries as well. But what happens is you need to realize that to support a foundry, there’s a thousand companies that are required to wrap around, build the shell, bring in the equipment, put up power plants, put up water reclaim apparatus, and then all of the different… Whether it’s the fobs going through with robotics, whether it’s all of the equipment-
Daniel Newman: Laser beams.
Patrick Moorhead: No, no. Totally. Two-story EUV laser beams from ASML for lithography or Applied Micro for backside power, Tokyo Electron. And every study that happened in New York said that it was a net gain. The tax on the workers and the companies that build this paid off in four years. When is the last thing spent by the government that paid for in four years? Name me one thing.
Daniel Newman: Oh, it won’t. I mean, look, every other part of the world is committed, and spending, and supportive, and we are dragging our feet. I don’t know why, Pat, we’re dragging our feet. I don’t get it. And by the way, these companies have fronted the money, they’ve moved forward with these projects they’ve committed to and now they’re waiting. And like I said, maybe they were overzealous, maybe they got too aggressive, but we need this here. You got to remember the basis of why the CHIPS Act became a thing, by the way. Remember it was resiliency, it was national security, and of course it was global technology leadership. If China invades Taiwan we are effed. We’re totally effed.
Patrick Moorhead: Totally. Yeah, I agree. One final comment that I’m mixed on, where does it go? FTC and the DOJ… I mean, Trump does not like big tech, particularly Google and I think Meta, and I just don’t think that there’s going to be no scars on that one when he’s done.
Daniel Newman: Yeah, I got super mixed feelings too on that one because if he takes them down, they weight so heavy on averages, the legacy… There’s that conflict of basically forming the market but doing what he believes. And I think he’s at the very least going to push them to fix the algorithms.
Patrick Moorhead: Let me tell you why I don’t think he touches Microsoft or AWS is national security, where I think that Google and Meta are not necessarily… Interestingly enough, Meta did the national security thing. Is there a correlation? I don’t know.
Daniel Newman: Will Apple continue to go unscathed? That’s another… God. The most obvious antitrust case of all of them is the damn-
Patrick Moorhead: Totally. Try to put their suppliers out of business, they don’t pay them, they lay off people. Trump had to come in and bail out Qualcomm. Qualcomm the national treasure, that’s what they called them.
Daniel Newman: Well, if you’re using a phone, you’re probably using some Qualcomm.
Patrick Moorhead: Totally. All right, let’s go. That was fun. That was fun.
Daniel Newman: Yeah, we should do more of that.
Patrick Moorhead: So let’s go to a more grounded one and simpler one. And that is that Solidigm, which is a company owned by Hynix, announced 122-terabyte SSD. The Six Five was there, I was there. We not only did it on stage, Melody Brue was there, but we captured it as well and I’m hoping, any day now, we’ll be able to publish that. But, Dan, you know I like to wax and wane on history and stuff like that, but-
Daniel Newman: Nap, I’ll come back in a few.
Patrick Moorhead: No, no. I want to try to put this in context. For AI, what have we been talking about? We’ve been talking about GPUs, HBM3, and silicon photonics. We have not really been talking a whole lot about storage. Maybe with the exception of VAST Data or something like that in relation to xAI-
Daniel Newman: Data infrastructure.
Patrick Moorhead: Yeah, embedding meta tags, stuff like that, databases. But here, we have storage. Right? And every inflection point, we screw this up, whether it’s mainframe to mini, mini to client’s server, social/local/mobile, web, e-commerce, AI version 1.0, it all gets crazy. And if storage, compute, memory, and networking and auto aligned, we do unnatural things and it’s super inefficient. So with bringing this 122-terabyte SSD, and a lot of people might not know this, but about 80% of all hyperscaler data is sitting on hard drives. Okay? Hard drives that are slow, they’re cheaper, that are not dense, that are power hungry and aren’t really meeting the needs of any modern application. Just to be clear, AI is all SSD.
So, essentially, Solidigm brought one of the solutions to the challenge of AI, and we always talk about data management, but I also don’t want to wax over storage because on top of every good data management platform, you have a ton of data and that’s why folks on stage with me and Mel was VAST Data CEO, Renen Hallak, and Arm executive, Ocient, CoreWeave right? The big front runner. They just got a big valuation, company called PEAKAIO that they’ve seen a 400% increase in business from AI on the edge. But it was a good showing, it was good thought leadership. It’s good to see a storage company for the enterprise actually do thought leadership as opposed to not say anything and duck and cover. So it was good stuff.
Daniel Newman: Yeah, I don’t have a lot to add there, but I did find when we went out there, we spent some time with our co-CEO and the team that they’re definitely doing some interesting things and there’s a tiering to all this. It’s not all just compute memory. The data needs to be stored somewhere. So the way that gets done and the way data is being tagged and metadata is being created and the way that’s being accessed by applications and the impact of agents is all going to be really, really profound. But in the end, this stuff still has to get done and they’re a well-run company with some really good innovation and thanks for taking that trip for The Six Five.
Patrick Moorhead: You got it, buddy. Middle-
Daniel Newman: You want to close this one out? You want to close this one out?
Patrick Moorhead: Let’s close this one out. The final thing I want to say is, on the same day, Micron came out with a storage solution that was half the size, but it was PCIe 5.0. But anyways, good to see thought leadership out here. Let’s dive into Cisco’s Q1 fiscal ’25 earnings. So had a great sit down with their CFO. That was a good to get, the play-by-play. What I’m going to talk about is my opinion. Right? So first off, beat, beat and a raise, and man, that stock went down. So all I can figure out is a selling on the news here-
Daniel Newman: I got one thing just to put up. They are actually still down revenue year-on-year.
Patrick Moorhead: No, 100%-
Daniel Newman: Beating expectations, but that’s probably at least somewhat influential in that.
Patrick Moorhead: It’s funny though, but beating expectations is always where it’s at. So anyways, I have a hard time hearing this one out. One thing I did appreciate here is it’s been a mystery sometimes to figure out what Cisco’s AI play is and how it translates financially. I really appreciated they had a slide in the deck that was crystal clear on where it plays. AI training infrastructure, AI connectivity, and AI inference out there and they listed orders, platforms, and outlook. Thank you, Scott and team, for doing that. And even better is they put a number, right? They cited the AI number at $300 million in revenue and where it would go over overall in fiscal year ’25, which was a billion dollars. If I look broad based and come out and I’m like, “Hey, what makes data go around?” Right? It’s compute, it’s data, it’s security, and it’s networking.
And Cisco is hitting on those three for sure, right? You have Splunk with the data, you have security-as-a-service or security appliances and hardware, and obviously the networking is obvious. Interestingly enough, they pointed out a lot of hyperscaler networking revenue, which I think always… We sometimes think of Cisco as an enterprise network and play only, but it’s good to see them bring that out and they do their own silicon with Silicon One. And the only reason I’m not… Even though the company came out and said they were all in on AI computing at their partner summit, I haven’t seen enough to think if it’s going to be material to the rest of the business. UCS and these AI pods seem like a really good easy button, and UCS has always been a good easy button when you wrap compute, storage, and networking in there, but margin structure on compute is a lot lower than any of their other businesses. So they’re not in the same business of pushing compute like Dell, and Lenovo, and HPE.
Daniel Newman: Thanks for not saying Supermicro.
Patrick Moorhead: Yeah, they definitely-
Daniel Newman: I think we didn’t actually do a segment on that. Can I just do one real quickly?
Patrick Moorhead: Please.
Daniel Newman: I think they’re getting delisted. I think that might be the best short of all time, or one of them. I think they’re the next Enron. Anyway, I mean, everything, I’m reading. I got no inside there. Pat, I think you covered that really well. I mentioned in the beginning they are in this return to growth. I mean, they’ve got great accretive growth with Splunk, 50% recurring software revenue now. They’ve done a great job diversifying, getting off this box volume strategy that it’s so long been well known for. I think entering compute again is going to be really interesting. I think you talked about the margin structure, Pat. I think picking their spots is going to be so important that they don’t try to attack it at volume. I think they need to look at where they can really add value, which goes back to their pod approach. For the longest time, it was always the ultra premium, highly integrated.
But, Pat, I mean, networking. Okay? Think about this. This is a company that emits this super boon and what does this say? This says that the enterprise is still in its infancy because NVIDIA has got the hyperscalers, and then some of this, the Dell and stuff is selling to the tier-twos, but where’s all the opportunity for AI? It’s going to be these on-prem smaller cluster AI data centers that are going to be used for large enterprises that either need to for sovereignty, data residency, lineage, and all these different reasons, compliance, or it’s going to be companies that are going to want to do a little bit more of their AI on-prem. They’re going to want to maybe get cost structures in place. It was still really early days to figure out what the long-term cost implications are for companies to run their compute. And there’s going to be a technology economics experiment that’s going to need to be run over the next few years.
Because technology’s going to become one of the biggest cost centers to companies to figure out how to manage this, how to optimize this, and how to get compute cost per… Not just token, but just cost of compute down to the lowest level it needs to be to run the business is going to be super interesting. And this is where Cisco should be able to shine. This is where they play particularly well. So I think it’s interesting they’re getting into compute. I think Splunk and the ARR is really compelling. I mean, but not seeing the networking grow faster amidst all the… You and I talk about networking for AI is because networking for AI is really still only happening in the tier-one, tier-two data centers. It isn’t happening yet at scale in the enterprise. And that, I think, is where… And we got some of that from G2 and the team when we met them at the partner summit, but I think that’s when Cisco starts to become a real beneficiary of the AI tailwind. That’s going to be a longer term, by the way.
Patrick Moorhead: Hey, have you guys done a forecast at Futurum Group data on when it hits?
Daniel Newman: Enterprise?
Patrick Moorhead: Yeah, enterprise-
Daniel Newman: I think we are working on data center infrastructure as we speak. But, I mean, right now, most of the category and most of the inputs… Now we’ll have some new IT. We actually have a CIO survey going out right now, about 700 CIOs that had control over $16 billion of spend and we’ll have their insights and we’ll be publishing that pretty soon. As soon as we do, we’ll talk about it a little bit here on the show. But, yeah, it’d be very interesting to see the longer tail. I actually talked to Dr. Howard Rubin yesterday on The Six Five Main Scoop, and he does technology economics and that’s what he does. And he and I had a little debate on the air, but he basically told me enterprises aren’t in on AI at all. He said from a numbers standpoint, in terms of their IT infrastructure-
Patrick Moorhead: That’s right. They did science projects and they’re doing some POCs, but they’re not scaling. I think it’s likely 18 months before we see the hockey stick. The biggest challenge right now is data management and securing that data. And I think that the land of models is going to be smaller, either horizontal or vertical models that are achieving different types of things, very focused on the actual business output and business benefits.
Daniel Newman: Man, what a show.
Patrick Moorhead: Okay, I guess that means we’re done. I love that.
Daniel Newman: I got to go to dinner. I go-
Patrick Moorhead: No, no. Hey, I want to thank everybody for tuning in. And even though Dan’s leaving, he thanks you. But, no, I appreciate you hanging around. I mean, Dan, we didn’t even go an hour and we cracked through nine topics, so I want to thank-
Daniel Newman: You know what I think the conclusion is here, and I don’t know, we’ll see how the show performs, you and I are less long-winded when we get closer to the end of the day.
Patrick Moorhead: I still have one more meeting to go.
Daniel Newman: I have a dinner. Yeah, I’m heading to a dinner. I’m just laughing because I don’t know that we didn’t say everything we should, but we maybe just didn’t say all the extra stuff that we tend to… Maybe at 9:00 AM on a Friday when we’re all jacked up on coffee and post-workout drinks.
Patrick Moorhead: I know. Totally.
Daniel Newman: By the way, any new PRs? Any new PBs? Personal best, personal records?
Patrick Moorhead: I didn’t injure my shoulder on flat bench. My trainer allowed me to go back for flat. I mean, I’m still doing tiny little weights, but I’m feeling it that… The seated row, I’m all but two plates on the full stack. I’m feeling pretty good about that. But, anyways, Dan and I are going to be spending pretty much all next week in Chicago at a Microsoft event. Very, very interested in seeing the state of enterprise AI. And when you’re Microsoft, you have IaaS, you have PaaS, you have SaaS, you do it your own, you have, “Leave the driving to us,” you have PCs with AI models. I mean, if there’s one company that pretty much does everything in AI, it’s Microsoft.
Daniel Newman: And, Pat, can I say, you know what else next week is?
Patrick Moorhead: What?
Daniel Newman: Happy NVIDIA week. I think there’s a whole week now. They’re-
Patrick Moorhead: Oh, it is. That’s right. You and I are both on shows. Yeah, I’ve got CNBC, I’ve got Yahoo Finance, and who knows what other trouble I can get on-
Daniel Newman: Just saying that’s next week too, so it’s going to be a banger. All right.
Patrick Moorhead: All right. Take care, everybody. Thanks for tuning in. Hit that subscribe button. Adios.
Author Information
Daniel is the CEO of The Futurum Group. Living his life at the intersection of people and technology, Daniel works with the world’s largest technology brands exploring Digital Transformation and how it is influencing the enterprise.
From the leading edge of AI to global technology policy, Daniel makes the connections between business, people and tech that are required for companies to benefit most from their technology investments. Daniel is a top 5 globally ranked industry analyst and his ideas are regularly cited or shared in television appearances by CNBC, Bloomberg, Wall Street Journal and hundreds of other sites around the world.
A 7x Best-Selling Author including his most recent book “Human/Machine.” Daniel is also a Forbes and MarketWatch (Dow Jones) contributor.
An MBA and Former Graduate Adjunct Faculty, Daniel is an Austin Texas transplant after 40 years in Chicago. His speaking takes him around the world each year as he shares his vision of the role technology will play in our future.
