 Hello everyone, welcome to episode 10 of theCUBE Pod. I'm John Furrier, Dave Vellante. This is our podcast weekly where we go break down all the stories we're watching and tracking interviews we've done, summary of where theCUBE's been, but really breaking down the enterprise and the overall technology business from Silicon chips all the way up to what's going on in the government. We'll break down the top stories. And of course, the famous rant section. This is our 10th episode, Dave. You know, we said we'd do 10 to kind of get the bait to go and get our group swing, get the format down. We're going to invite guests and we had John Chambers on last time and we're at 10. Welcome. Yeah, thanks John. I look forward to these every week now. You two really scan the news, give our opinion. And it's great to be here. It's excellent. I don't think we're going to have any Elon Musk session this week, but we do have AI backed, headlining. We've got tons of great stories to review. We're at AI alarm bells are ringing louder than ever. The skeptics are coming out. We've got a rant section because Lena Khan, your favorite person, our favorite person, but you like her more than I do is in the news rather than opted with New York Times around regulating AI, which is ridiculous. John Chambers last week on the podcast went out of other direction. Other stories, you know, Rob Hove and the team are looking at all the sentiment and with the influencers in the industry around AI, content pollution. Vice wrote an article. They filed for bankruptcy. I said, you know, that media inclusion. We talked about that last week. Again, JP Morgan bought acquired first Republic. Again, another bank failure continuing to happen. For a song. Our podcast is just triggering all these ads. Our first podcast, everything's happening like over and over again. IPOs might not be dead. And the PC market is hitting a bottom and a rebound is incited according to Morgan Stanley. We'll unpack that security. We were just at RSA. So much going on around the industry. Dave, again, you know, the Godfather of AI. Google quits. Just so much happening around AI. Just the AI stuff keeps going. Security problems keep cropping up. We're at RSA last week. We're going to be in the open source summit in Vancouver. I mean, it's crazy, Dave. AI never stops pumping out great content. Hype, meat on the bone. It's just a cycle of AI. I think this is going to turn into an AI podcast. Yeah, and we had rate hikes this week. The Fed raised rates to five and a half percent. Oil is down, you know, to get under price pressure, which I think surprised some people. And then the jobs report today comes out. And unemployment is like going down. And yet the stock market rockets. So earnings, earnings, crazy numbers, crazy numbers on the earnings reports. Maybe Apple, maybe, maybe because Apple, I mean, Apple did, you did better than expected. Even though its revenues dropped, I mean, it threw off, you know, amazing numbers. A $90 billion stock buyback. So that's fueling the market. I mean, $90 billion. The numbers are just mind boggling when you look at Apple and their balance sheet. I mean, it's just insane. Arm, arm is filed to go public in this market. What? Oh my God. You remember, NVIDIA tried to buy arm for $40 billion, but the UK probably would lean a con whispering in their ear, you know, killed that deal. And so the US could have had another just awesome, you know, ownership of a great company. So now they got to go public because, you know, they got no other options. You know, arm is really interesting business model. They don't make anything, but they do produce, you know, awesome frameworks and software. I don't, I mean, they don't make any silicon chips, but everybody designs chips, not everybody, but many do design chips using the arm format and then they toss them over to TSMC. It's really Apple, arm and TSMC are really driving innovation and semiconductors, particularly in volume and everybody's sort of chasing that, that next AI inferencing at the edge and just can't wait to see that S1 and that registration. It's confidential. We can't read it right yet, but we'll see. The rumors are falling all around. Even in the Silicon Wars, I mean, Amazon and a Pernod acquisition, there's rumors I'm chasing a story down right now that AWS might be exiting one of their chip businesses. So that they were made to have to go with a supplier. Yeah, so I got a hot lead on that story. Yeah, Amazon doesn't want to sign up with the big company. So they want to make their own. They can't always do it. Well, you know, what's interesting is, I mean, I remember reading something at Hawk Tan said a while back is basically on an Ernie's call saying the market's shifting toward merchant silicon. And he's right generally, and the exceptions have been the big hyperscale cloud providers. That would be big news if Amazon decided to back off. What can you share on that? I really don't want to get into it. I have good sources telling me I don't want to reveal it now, but you mentioned Hawk Tan at Broadcom. That's a company that's doing extremely well, Dave. I mean, I've been following the Broadcom for years and when they made that move in there, they have such a great supply chain right now and they have great customer OEM base. They are the leading player. And if you look at their stock, they're huge. If you look at Broadcom as a company, I know you did a post on them a while back, but if you look at what they've done, and I've really looked under the hub cover since they acquired VMware or tried to close it yet, but the VMware acquisition, which we've analyzed till the cows come home, but you look outside the VMware when Broadcom is done, they bought CA and integrated that, but their chip business is significant. And remember, we are covering of the cube and silicon angle. We set the narrative on the architecture. Super computing is going through a renaissance. Super cloud is emerging and super apps are here. It's a super cool market, Dave. I mean, this is the super infrastructure going on at the lower levels is mind blowing. And Broadcom has a huge edge there on all aspects of networking and chips. So a little secret there. Well, you know, it's interesting about Broadcom is they're kind of like the Coke industries of semiconductors, right? They're not really the sexiest company out there, but they just, they buy up assets that on their own didn't look that interesting, but then they have the awesome supply chain. If you look at their stock, like since late last year, right before the market tank, they didn't, they just kept cruising on. They throw off a ton of cash. And so what's interesting is their basic premise is the world is shifting from a processor-centric world to a connect-centric world. And that's, that's cause they don't really play in processors per se, but it's kind of true. All these alternative processing engines and capabilities, whether it's network interface cards, RAID controllers, other connectors are becoming increasingly important. And it's just unbelievable to see what that company is. And by the way, 50% after the VMware acquisition, if these knuckleheads let it go through, is gonna be in software, 50% of their revenue is roughly will be in software. And the premise is, oh well, it's gonna hurt competition. I'll tell you right now, everywhere I go when you talk to customers about that acquisition, they go, yeah, we're checking out some alternatives. We're looking at Red Hat, we're looking at Nutanix. So if anything, it's gonna increase competition. It's just, that would be stupid if they killed that one too. Dave, well Dave, super compute, super cloud, super apps. That's the three layer stack that we're reporting and defining, framing that in the industry. Out in the open, by the way, no one else is doing this. We're doing it theCUBE and SiliconANGLE. But let me tell you why I like Broadcom. If you go look at what they've done since 2012, 2013, I call that the major inflection point of AWS. If you remember that time, the luxury of theCUBE being 13 years old is, we were around in 2010. We've seen everything, right? So I can tell you, this is more historical view. 2013, everyone thought cloud was just a piece of shit, right? They didn't, ah, cloud that's missing the junk yards. Do-it-yourself, do-it-yourself cloud. Okay, if you look at 2013, okay, let's pick 2014 as an inflection point because 2013, the world wigs up the cloud. 2014, Broadcom stocks at 45, okay? You map Amazon stock to Broadcom. You see a trajectory, okay? On December 31st, 2021, high watermark for Broadcom was 665, okay? It's 627 right now, okay? They haven't moved, they dropped, they dropped in October and they're back up, okay? So October 2022, okay, they're a beast. They're quiet and they have the super compute angle. They're gonna power super cloud and ultimately the apps are gonna run on it. And then you have this other market developing, I call the Bolt-ons. You got the, what GPUs used to be just for graphics? No, that's powering machine learning. So, quantum's around the corner. You got VMware control plane. They'll probably buy some container Kubernetes company control that layer too. So, Broadcom Dave is a beast. They are killing it. They're doing great financially. So, their strategy, whatever they did in 2013, 2014, has been working. The shift in cloud operations, shift to higher performance silicon. We just wrote about this. When we were talking about the, the semiconductors were leading, are typically a leading indicator of a rally. We had Ivana Dilevska on breaking analysis and she was talking about how she thinks that it's a harbinger to good things to come for tech. It's still weird right now though. The economy's very difficult, right? People don't have great visibility. Although, I think what's happening is that smart CEOs and CFOs in tech were really conservative. And so, what's happening is earnings estimates came down and now they're beating very conservative earnings estimates. So, this could be a little bit of fool's gold but with generative AI, the data center is going to explode. I mean, there's no question. Is there any question in your mind that this AI stuff is going to drive like tons and tons of hardware or software demand? Zero doubt in my mind. In fact, I'm over the moon on the AI push. AI will create renaissance in every single market. Like I said, super compute, super cloud, super apps, the edge is coming. When edge develops, edge computing, that'll put the final pillar for distributed computing and then an OS will develop. And then that layer will be full multi-cloud, full super cloud and then everything will matter. Performance, hardware is coming back. Morgan Stanley just put out a report that the PC is hitting a 30 year bottom. Think about that. Okay, I just prepared the Broadcom stock to what it is now. They're expecting to come back because in every cycle there's going to be a refresh. The PC is going to turn into an intelligent device. Like everything else, like the phone became a computer with the iPhone. So, I think you're going to see hardware matters. We've been talking about that. Speeds and feeds and apps. Nobody wants to run their application on slow infrastructure. So, no one wants that. So, you're going to see a huge race for who's got the best hardware powering it. And that hardware, by the way, will be in your pocket, on your body, in the data center, in the cloud, in a car, everywhere. So, you're going to see a whole new connective tissue of software and the generative AI hype, to me puts another paradigm shift into the equation because ChatGPT just introduced the notion of large language models and large multimodal media, audio and video, multi-media foundational models. That didn't exist in the context it did five years ago at cloud. So, that's what's different about AI is that you now have data doing it. I'll give you an example. You're going to like this. I just did an interview with some FinTech gurus. The old Wall Street trading model was frequency trading. Every packet would move through the wire. Every nanosecond you get advantage, you have a better trade advantage on Wall Street. Makes sense, right? You get the data first, you see it first, you trade on that data. You ever read that book, Flash Boys? Flash, exactly, yes. Michael Lewis, AI frequency trading. They put the data centers right next to the trading floor is a little of latency. Every squeeze, every ounce of performance out of the packet. If you haven't read that book, you should pick it up. It's called Flash Boys. Michael Lewis is just he's the guy who wrote Moneyball. And basically the traders, they would pay like millions of dollars, exactly what you're saying, John, to put, I want to put my server next to your server. Like literally physically, I want it right there. So they're jockeying for tech position? Yeah, for sure. Okay, so now here's what's different. We just riffed on this on an interview. The new model is getting data advantage, not packet speed, but having the right tuning fork, if you will, for the, for the data, by using the large language models, by putting data in the prompts and tuning it. So the new advantages, prompting data sets to produce insights faster, which means it's like whoever can get the right incantation of the data, boom, magic, you know, like Harry Potter, Exploriamus, that's the new advantage. And you could trade on that because the psychology of like say, the marketplace will get that insight as a market moving up opportunity. So it's first, get the packets with physics, packet moves from eight point, eight point B, make that faster. Now the advantage is data insight acquisition. What data can you inject into the AI to spit out the answer? First, that's going to be a game changer. And so Wall Street's already onto this. So they like building their language models and then tapping the big ones and just tuning, almost like probing, almost like a Google search. I have the right keyword kind of thing. It's really fascinating. That's the new advantage. It's not physics, it's data. And so that to me is going to drive infrastructure compute. That means they're going to buy more service from Dell. They're going to buy more service from HP. They're going to buy more cloud compute. So compute will drive that behavior. So this week in breaking analysis, the title is Desperately Seeking Cloud Repatriation. And the catalyst for this was 37 signals is, you know, leaving the cloud and why they, you know, they wrote about why they're departing the cloud. And so looked into a little bit. But one of the things that I sort of thought about pretty deeply and, you know, we've been doing a lot of work in the HPC supercomputer space is, and I'm trying to figure out exactly where open AI servers live. We know they're in Azure, but they train their models on supercomputers in, I believe it's Ohio. And I asked GPT where the servers are and it wouldn't tell me. And I asked it, is that confidential? And it sort of hedged and bobbed and weaved. But I found some stuff on Reddit that suggests that there are supercomputers in Ohio. So it actually kind of makes sense, John, supercomputers at scale. If you have more than 50,000 cores and some of these supercomputers have millions of cores, it actually makes economic sense to have them on-prem. As well, you're going to do AI inferencing at the edge. We talk about that all the time. So the point, the reason I'm bringing this up is will generative AI sort of change the balance in cloud versus on-prem? We know there's a lot of stuff going to be done in the cloud. That's where the training is going to be done for sure. But the edge and high performance computing might change that balance a little bit. And I was looking at one of my things I references is our friend, Charles Fitzgerald Fitzsie's repatriation index, where he basically takes digital realty and equinex and divides it by their revenue by AWS revenue. And it looks to me like it's been declining for years. It looks to me like it's flattening and going to pop back up sort of favoring on-prem a little bit. And I've been so negative on repatriation because I think it's just bullshit, but maybe there's something to it. Well, I mean, I think repatriation is the definition that needs to be looked at. I don't think people are repatriating in the pejorative sense of, oh yeah, cloud's not working. I think they're repatriating or just re-architecting their platform to be cloud operations. So if you look at the data centers, and take the data equinex example, that's not so much about data centers as more as multi-clouds. So for example, the cloud providers are still the big buyers of data centers. So Amazon and Azure are going to be constantly and Google rolling up data centers, even Oracle, their core strategies that own their own data centers. So their cloud is a bunch of data centers. So the data center market is moving away from managed services to be bought by Google, kind of bought by Amazon. So the hyperscales I think will keep buying capex. And that's going to be a signal that multi-cloud is probably going to be more legit. And then I think the data center for enterprise will have their proprietary stuff on premise. As long as it's running cloud operations, why wouldn't you keep your proprietary stuff online? The only thing stopping that's the networking. I think cloud native networking is a problem. Anyone who's doing hybrid cloud right now will tell you that the hardcore networking is not yet cloud native. There's different networking stacks in Amazon cloud than there is in premise. So stay on that for a second. The cloud operating model, if you will. So Laurie McVitty wrote a post talking about the relationship between SRE ops and on on-prem repatriation, whatever you want to call it. And it aligned with the 37 signals. And basically their point was, look, 80 to 90% of the skills that we need to run in the cloud, we have now on-prem. We don't have to do FinOps and forensic accounting on our cloud bill. We don't have to do that. And yeah, we got to replace disk drives when they break. But the cloud operating model is moving on-prem. It was interesting. You were a sit-down with Adam Silebski last year. He said, no, that's not cloud on-prem. And of course the on-prem guys say, not everything's going to the cloud. So we're starting to see this a bit of an equilibrium. Also Andy Jassy, you said in the last earnings call, remember 90% of the workloads are still on-prem. Well, it's been 90% for the last three or four years. So it can't still be 90%. So things got to be changing here. And I actually think we are starting to see a little bit of a tipping point here. Maybe not a tipping point, but a little bit of an equilibrium reverting to- Well, I mean, I think enterprises are settling into the value proposition of cloud, of the public cloud. And hybrid's been dominating for the past, half a decade in up in production. I think people are seeing where the gaps are in the pandemic pulled forwards in issues. Firewalls, and for instance, the next gen firewalls will probably be distributed. We've been talking about that privately behind the scenes. We'll have some reports on that soon coming out. What Palo Alto Networks is doing with the platform security. There's a lot of issues. So it's not as easy as the lift and shift to the cloud and just move all your data centers. As Andy Jassy used to say, oh yeah, just migrate everything to the cloud. Yeah, you can burn the boats as they say in that metaphor and go all into the cloud and deal with it like the capital ones of the world, but most enterprises can't do that. They got security issues, they got operational challenges, and they got to stay on premises. So that's to me the big core. And I think AI will force them either to be more on-prem or more on the cloud, depending on how the cloud guys come into it. What Amazon is releasing, what Microsoft has, what Google has, are potential game changes to keep the customers in the cloud while making the sovereignty of their data and having global expansion fast. So it's going to come down to architecture. Last week in the pod, we had John Chambers on CEO of Cisco. He basically said in context of security, but it provides to everything. It's an architectural game right now. Whoever architects their system, their company, I won't call it IT, but their company technology platform will win. Whoever does that the best wins. And if you're an architect, you're thinking, okay, what's the blueprint of a great enterprise, global business in today's current state of the market? And it ain't what it was. Okay, that's a fact. It ain't what it was 10 years ago, or 20 years ago. It's going to be completely different. And that's the challenge. So you remember, so doing a CUBE pod with you, it's like when you ask questions of guests, like remember John Chambers when he came on last week, he was like, well, John, there's like six questions in there. I'll start, I'll try to answer them. But so I want to go back to something you said. So you remember, so you're talking about lift and shift. This was like three points ago. Remember Alan Nance, our friend, who at the time was the CIO of Phillips, who's CIO of like some division of Phillips? He said to us like years and years ago, look, if you lift and shift, yeah, you get a couple of benefits, but if you really want to drop numbers like zeros, multiple zeros to the bottom line, you have to change your operating model. And he was clear on that. Like this is like six or seven years ago. So that's proven true. And then the other thing is the guys at 37 Signals kind of picked up on something that VMware has been saying more. They used the term sovereign cloud. They were proposing that as opposed to private cloud or saying rental, which is rental, cloud is rental. But sovereign cloud, not something that, VMware has been using that term, I think since 2018, our colleague Dave Nicholson always talks about this. He says, oh, it's just IT. And I used to say, well, it's really not just IT, but again, I think there is this equilibrium. Well, you're both right. It's the new IT, right? Yeah, exactly. And what was IT? Again, we've said this in so many Q but interviews. IT was there to serve the business, provide technology to the workers and meet the competitive advantage of the company, build the website. That was small and narrow. Now every company is a full blown technology company, run by technology, powered by technology, powered by servers and data and applications because their customers will be interfacing with them in some capacity. So to me, that's the real issue going on now is, what does that look like? That's why I say the new IT is just technology. That's in every aspect of the business. So the architectural question is, throw out the old building and rewrite new blueprints, like you're architecting a building. You got to design it. And that's going to take brains and that's what the cloud brought to the table. And that's what AI is going to bring to the table. As this three super stack emerges, super compute, super cloud, super apps, that's the new architecture. And if anyone's not thinking about how to tune each piece, they're going to be out of business because AI is going to sit in all aspects of it. But at the top, it's going to be apps going to have to thread through those three layers. So every time I go to Silicon Valley and because you live there, so you do this every day, but every time I go, I learn something that's like makes me think differently. And I want to say it was like five or six years ago I was there and I was talking to one of these alpha geeks and the individual said, look, doing this with our thumbs, that's not the way we're going to interact with technology in the future. That's like to say five or six years ago. And so now we're seeing that just as Amazon turned the data center into an API, foundation models like GPT are going to turn technology into natural language interfaces. So like, I've been thinking about this in terms of, one of the sort of downsides of mobile and all these wonderful apps is I got too many apps. Updating them all the time, forgetting your passwords. Every time you change your phone, it sort of doesn't work the same way. If you use non-SMS based two-party, two-factor authentication, you got to that they don't support them all. It's just a real pain in the ass. So my question to you would be, how do you see foundation models, GPT and AI, sort of changing that mobile, typing with your thumbs or really, with your fingers with a zillion apps that are all sort of these bespoke apps. How do you see that paradigm changing as a result of AI? That's a loaded question. I'll just come back to my standard response. Everything that will be invented in the future was on Star Trek or Star Wars. Except for the transporter room. Physics can transport people from point A to point B. It's going to be voice and gestures or whatever. You're starting to see that now. Virtual reality comes in. John Chambers brought this up on the pod. Yes, again, another John Chambers reference. He said voice is going to be actually identifiable, not just unique tones with people's voices. So we're already doing that now. Hey Siri, hey Google, do this, the kind of thing. So I think you'll see AI become autonomous at some point. The large language models that we see today are scratching the surface. They're going to be prompting. First you get the prompts down, that querying data sets, merging data together, throwing data at data. It's a chemical reaction, that vibe. Once that is done, you operationalize the data. That's what we've been covering for 10 years. How do you bring data into an organization? Keep it safe, protected, but make it addressable, let developers integrate it into applications. Once you operationalize data, then you get into the fine tuning the data. I call it that the tuning fork, for like an instrument that you tune the data. And then once you do that, it's trained to self heal and be autonomous. That's the trajectory of where I see AI going. We're now in the tuning of prompt stage. Prompting, chat, CPT is an elementary example. Okay, so operationalizing it's going to be like, okay, another complicated thing for an enterprise to do when they're doing a rearchitect. Imagine doing the blueprint for a big building and someone invented elevators. Well, I didn't think about that. So like, okay. Oh, right. It's like, all right. Exactly. Shit, we fucked that up. Okay, go back to the drawing board. But that's what's happening right now for every enterprise with technology. Okay, so you think about how we use apps today. You mean think about the apps you use. You use Waze, all right, summer's coming. So you're going to break out the site. Well, on the East coast anyway, we're going to break out the road bike. So we use Cyclometer or whatever cycling app you use. Whatever app it is, you're trying to find a level. So you use your phone app with the level app because you want to hang a picture or whatever it is. So today you find the app, you load it up, you configure it with your thumbs and then you use it. So in the future, it's going to be different, right? You're going to say, there's where I'm going or this is the route I'm taking. You just fire up the route, boom. Or track my ride. And it's going to be the ride three or whatever you call your ride three, up Bear Hill Road. And it's just going to happen on this device, right? Do you see that? So what happens to all the apps? What happens to the app store? The app store just feels outdated to me. And same thing with all my streaming services. I don't even know what streaming services I pay for anymore. I'd like, do we have Hulu? Yeah, do we have Hulu without ads? I just tried to log into a service and they gave me the email option or Google. And I signed up and I paid for it. I see it in my credit card. I'm like, okay, I want to use the service. I try to log in. So you want to have to sign in because I logged in with my Google credentials. I owe up to the Google single sign on. But then I'm thinking, shit, did I do it with Google? And like, no, I didn't. I did it with the email. And I forgot the password. Then it says send an email. I was like, I can't find it. Hey, yeah, he's got to solve this problem. And it's a technology is just so much. Well, we have too many things, right? We have too many subscriptions to service, streaming services. We got too many subscriptions to sub stacks. I got too many subscriptions to websites. I got too many apps on my phone. I think what's gonna happen is you're gonna see AI bring personalization to a level that we've never seen before. They're gonna, it's gonna know what I need before I need it. You already see that on the web with Facebook. They already got the hive of mind going on where it's like, they know they show you ads but it's mostly retargeting. There's no intelligence to it. But the mechanisms are in place to query an entity, a being, a collective software mechanism that says, hey, assist me. That's why I like this co-pilot vibe going on. Help me out with recommendations. You can save my iPhone. Kill all the apps I haven't used in the past six months. Delete and unsubscribe. Hey Siri, it won't do that. That will happen. That will have to happen. Someone will do that. That's an entrepreneur. Someone out there listening probably in high school and college do a hackathon. They build that app. It goes viral. Done, that they're in business. That's the entrepreneurship equation that's changing. It's so exciting is that it's gonna be the super compute, super cloud, super app stack. The super stack model is legit game changer for entrepreneurs. New brands will emerge. Have you written code yet in chat GPT? No, I haven't. But our team has, right? I mean, I haven't either. I haven't either. I mean, yeah, you have, but you got to test it. You roll it out. If you're taking it around, you can write, have it write a website for you. I had it, I had it to give me a outline for a product plan for an e-commerce site. It's spit out essentially pseudo code and basically the site heuristic. It's there, it was all there. Yeah, so like the reason I ask is like, you think about, everybody talks about democratizing data, citizen data scientists, citizen developers, and you couldn't really see the path to get there. I couldn't, but I think the path is pretty clear now to the extent that the stuff can actually work and not be, not give crap. I mean, like, at one point I asked, I was ego GPTing and I was, I had worked for the Wall Street Journal, which yeah, it's a nice compliment, but not true. So you wonder, does it code better than some of the responses that it gives? I don't know. I mean, I don't, I'm not high on the whole coding thing. I think it'll be more of a write code for me and then I clean it up just like with writing, right? You write something. Dave, so like, not to change gears, AI is gonna dominate everything. One AI point before we go is there was an article on Vice, by the way, Vice just going into bankruptcy and it looks like they're gonna get bought out by 400 million, I saw a Wall Street Journal article today on that. But they put an article about content pollution and it says AI is tearing Wikipedia apart. So we, I bring this up because we brought this up two pods ago at KubeCon, which is the cloud native conference, the Linux foundation. I was talking about this with code that you're gonna see more code and more content pollution, more code that sucks and content that sucks come from AI because it's like not even good vanilla ice cream. It's like bad. So you're gonna see a lot of misinformation. So this article was the first one that I saw that called that this out specifically because Wikipedia, as you know, is a volunteer organization and it doesn't always have the staff to stay on top of the shit. So there's a lot of spewing of content hitting Wikipedia at such velocity that they can't keep up with it. So- And it's bots you're saying? Yeah, AI generators are just writing believable content that sounds credible. So it's harder for the volunteers to essentially edit it. So again, this is, we knew this was gonna happen. And again, the chat GPT and all the benefits of AI has a downside. So this is why everyone's freaking out about it because there's a good side is always a bad side. Remember when Facebook was launching, we were like, we said, we were like, there's a lot of tire targeted virality opportunities for content connecting. We were the first ones actually call out misinformation. It was like 2012, 2013. I remember where we were too. We were sitting in San Jose at like a big data of SV. Remember? Yeah. And you sort of laid out how misinformation- It could be manipulated. The payload is about misinformation. You talked about the weaponization of social media, which I think you were the first that I heard talking about that. And then- I think- So you predicted that accurately, obviously. I don't think you predicted the degree to which it occurred, but oh my God- Well, we were the first ones reporting it and no one sat there for years and then the election hit up, everyone freaked out. That's why I feel the same way about the super stack vision, super compute, super cloud, super apps, the super stack. So our coverage with super clouds continuing. You got our event coming up in July, July 18th around security, it'll be interesting. But anyway, AI is rocking. Let's transition to earnings, Dave. I know you're tracking a lot of earnings. One of the things I noticed this week was a slew of crazy earnings numbers. It was all over the map. Not sure what it adds up to, but AMD stock was down on 9% revenue decline. Qualcomm's stock sank like a rock, worsening more bad news to come as they predicted. Informatica's shares fell despite solid earnings. iPhone sales drive to solid earnings in a revenue beat. RM files of paperwork to go public? Is that good news? I was like, what the heck is going on in the market? What's going on in the market right now for public companies? Well, so let's start with AMD. AMD actually had a pretty decent corner. The stock dropped and I think it dropped. The, I guess people weren't excited about the outlook and PCs are sort of crummy, although the data center outlook to me looked pretty good. So the stock drops and then like the next day, it was announced that Microsoft was going to invest in a finance AMD's AI chip that Microsoft's going to use. And the stock goes back up. So on the one hand, you're like, okay, there's a lot of AI hype. But on the other hand, exactly what we were talking about earlier it underscores the degree to which data center is going to explode and hardware is going to explode as a result of all this generative AI. I think Qualcomm was a function of just tepid guidance. So maybe they're just being conservative. I mean, Apple's revenues declined for the first time, two quarters in a row. They're forecasting a decline and going forward. And yet the company beat its expectations and now it's the $90 billion stock buyback and its services business is I think holds a lot of promise. I think that's a lot of people see is the real potential for Apple going forward is not just handset, iPhone sales. You're now an entity at which they can continue to upsell whether it's three year maintenance or components or add-ons, PCs were dead. I mean, PCs declined, I want to say 30% or so the PC market's right in the tank. But I think you pointed out earlier off camera that I think it was JPMC. No, some Wall Street analysts said we've bottomed PC demand as- That was Morgan Stanley. It was Morgan Stanley that PC demand is bouncing off the bottom. I don't know, maybe I haven't seen it yet. It hasn't shown in my numbers, but so that's all happening. I am, I'll tell you, digging into the ETR data lately, some of the big names are definitely softening. I mean, Snowflake, one of my favorite to watch has come way down in terms of spending momentum. That's the sort of bad news. The good news is most of that is because it's shifting from we're spending more to our spending is flat. So it's not like they're spending less and the Delta, the very, very small numbers of people that are leaving the Snowflake platform, it's virtually non-existent, but nonetheless, the top end, those new ads and the percent of customers spending more is definitely trending down. So you're seeing that across the board. And I think that's a near-term phenomenon of optimizing cloud and taking advantage of savings plans. I was looking at the Snowflake 10K from March. They just checked this out. They just negotiated a new cloud contract between, I think, 20, starts this month or last month and goes out to 2028 where they're committing, get this, to spend 2.5 billion with a single cloud provider, ostensibly it's Amazon. Now, you might say, wow, there you go. Andreessen Horowitz is right. Martin Casado and Sarah Wang talking about, you know, the cloud repatriation and all the costs of goods sold, blah, blah, blah. But the other footnote, and I saw in the 10K was that just, they just essentially retired the previous commitment of $730 million. So what's happening, John, is they're basically doing what we've been talking about, the customers are doing, they're taking advantage of saving plans, committing more to the cloud and generally in AWS specifically, locking in longer-term contracts at lower prices and committing to an actual spend over some period of time. So to me, that's a really smart move by Amazon and the other cloud vendors to do that. And the last thing I'll share is the other data that I saw was, it was a survey on which cloud vendors are doing the most to help you. And it was in the order of priority was Google, Amazon and the furthest away was Microsoft. And that's just, I think a function of Google needs to gain share. Amazon's being really aggressive and Microsoft's got the software estate and it doesn't have to. Yeah, I mean, that's a great point. I have to ask you because this is, you know, when you have, we love to talk headwinds and tailwinds and love that when we get into that. You know, obviously we're in a headwind situation and during headwinds, you kind of optimize. And tailwinds, you expand, let your wings fly. How has the macro headwinds affected the expansion of or impact of existing customers versus new customers for Snowflake, Datadog, MongoDB, Confluence, you know, the public SaaS companies that you're following. I mean, how has it affected their growth for existing customers and new customers? What's the read there? What's the vibe? So what's happened is that the percent of new customers that are being added is definitely well off where it was at the sort of peak pandemic levels. So that's sort of point number one. There's a big shift toward, you know, getting revenue and incremental revenue out of existing customers. That's where they're putting their emphasis. But the vast majority of customers that I see in the data are sort of holding firm, keeping things flat, readjusting things, you know, maybe pairing back a little bit, and they're definitely not those names you mentioned, they're definitely not defecting. Those are good names. There's not a lot of churn in there. They're still in the leading companies. And I think they're positioning for when the market comes back, I think they're very well positioned there. So fewer new customers, but very few defections. What about the contrasting views of Amazon versus Microsoft regarding back to normalcy around optimization headwinds? Next few quarters, what implications does it have for the industry? Well, I mean, I think, you know, unlike most markets where it's winner take all or winner take most, you know, most markets, number one makes all the money, number two does okay, and number three barely breaks even and everybody else is crap. In this market, Amazon is, you know, they're number one in infrastructure as a service and Microsoft because it has such a huge software estate gets the captive infrastructure as a service base. And they're both like crushing it. Now, having said that, I think Microsoft has got more momentum right now and they're about 70% last quarter of Amazon's IaaS revenues, which is pretty astounding when you think about it. And then there's Google and the good news on Google, the bad news is they're way behind. They're tiny compared to Microsoft and Amazon. The good news is they're making a little bit of money from an operating profit basis and they got great tech and they can afford to play the long game. So you got a case where number one and number two are basically two number ones and the number two has a potential to be or number three rather has the, which is usually break even has the potential to be like a number two and make a little bit of money. And that's kind of how I see the big three cloud players. All right, let's get into the rant section. Enterprise is still kicking ass. I just think cloud and AI is going to be game changer. Again, the super stack and just coined that term now for lack of a better description, super compute, super cloud, super applications. And then the super edge will come online too. It's the super stack. That's going to be a focus architecturally. Now let's get into the rant section because if AI gets stunted at birth by regulation, your friend, your favorite person, Lena Khan, okay, is that at it again, Dave? She wrote an op-ed in the New York Times, not just ranting on like, you know, public high big platform, New York Times and her byline. She's saying, let's regulate AI. John Chambers, a legend in the industry who's been there, done that. Lena Khan just barely graduated high school. You got John Chambers and Lena Khan. John Chambers was like, why would you do that? It's the dumbest thing ever. This speaks to where we're at, Dave. So let me say this. So it's unbelievable to me that this article that she wrote, it's the same old, same old negativity, fear-mongering. She didn't even talk about the benefits. I mean, she just poo-pooed the benefits. She gave it sort of lip service, number one. Number two is I look at what's happening in crypto. The crypto community has been asking the government to go ahead, please regulate it. Give us some guidelines, give us some guardrails that said the government has given them nothing and now is attacking them. And my fear is the same thing's gonna happen here. Okay, you got some ideas on guardrails, put them forth and let's talk about them. Let's implement them. You know, Sam Altman was saying they met recently just yesterday with the government and it was a really productive meeting. Okay, fine, whatever. But you look at her op-ed here. We must regulate AI. Here's how. She didn't say, here's how. All she did is stoke fear, negativity. You should see, I mean, let me just share some of this stuff because it's driving me crazy. Public officials have a responsibility to ensure this hard-learned history doesn't repeat itself. She was talking about Web 2.0 is such a bad thing and how Google and all these free services are so bad because they surveil us. Well, I mean, there's a flip side of that where they're actually not so bad. And so, you know, yeah, there's definitely risks of AI. She talks about we need to maintain the open and fair competitive markets that have underpinned both breakthrough innovations in our nation's economic success without tolerating business models or practices involving the mass exploitation of their users. Okay, well, let me ask you a question. So don't you think you missed this already? Because Microsoft is a be a myth and OpenAI have got together and they're like out. They're like sprinting out in Kentucky Derby this weekend. You're gonna, you know, so the question is, are they cheap speed? I don't think Microsoft and OpenAI are cheap speed. I don't think they're gonna fade at the top of the stretch. So, okay, so you missed that. And so, I mean, what are we talking about here? And then the last thing I'll say, and then I'll shut up, just talking about this mass surveillance system. How about the finance companies, the banks, the credit card companies? Do you know, most people don't know this. When you sign a credit card, let's say you think at the teaser rate, 7% or 0% for a year, right? You see that all the time, 0% for six months. Do you know that they have so much information on you if you miss a payment outside of, let's say you do, you get a credit card from some bank, irrespective of whether you're on time, let's say you're on time payments with that bank every month, like clockwork. But if you miss a payment to your car payment, your mortgage payment, some other credit card, they have that information on you and they can bump up your rate from 0% or 7%, whatever the teaser rate is, to 22%. They're legally able to do that. So they have all the data. They have mass surveillance going on. Why isn't she talking about that? It's screwing consumers way more than free search or ways or bang. I mean, it's absurd. Yeah, that's a good rant. That was a good one this week. Well, the New York Times article totally catalyzed that rant and rightfully so. Here's my issue with her. First of all, I understand and I can appreciate where she's coming from. I just don't like her views and how she just, it's become very political. She signifies to me exactly what's wrong with our system. It's short-term political motivated, you know, virtue signaling optics. Nothing to do in substance. However, she does put some wraps yourselves around the great values of America at the end of her articles, she says, America's longstanding national commitment to fostering fair and open competition has been an essential part of what has made this nation an economic powerhouse and a laboratory of innovation. Yeah. Yes. Yeah, say, exactly. Let's hold that, put a pin in that. Put that to the top, post that tweet at the top, pin that tweet. That should be the North Star. And everything she says is completely opposite and she's using like web 2.0 as an example. I mean, first of all, the government didn't even know what web 2.0, the industry didn't even know what web 2.0 is. They're trying to say that Facebook and the drama that happened was because a web 2.0 wasn't regulated. That's complete horseshit. That's not true at all. Okay, that's complete garbage. But she just doesn't understand it. So here's my take. She needs to get a new set of advisors around her to actually know what they're talking about, have a historical perspective and actually understand what the American national commitment is around competition. Because what's happening with lawmakers like her at the FTC and everywhere else, lawmakers I'm saying, they don't know what game they're trying to regulate. That's like trying to like basically be an official NFL football, but it's really at the baseball rules. They don't know what the fields are changed. So the world's got to first understand what competition looks like. You know my saying, Dave, you can't play chess unless you know what checkmate looks like. So they got to get into the game and understand actually what they're regulating. You can't regulate what you don't understand or you can't even manage what you don't understand. So to me, this is why AI and this whole, this thing is BS. Yeah, guardrails, I'm cool with guardrails. And yeah, this is a political statement. Okay, small business, we're not gonna let AI hurt you. All that's just nonsense. Let the technology ride. John Chambers nailed it. It's a competitive global issue for the United States. The United States should lean into it and get in the game and figure out the game. Be in the arena. That's my view. That's what Silicon Valley should say. That's what everyone should be messaging to DC and lawmakers. Get in the game. I gotta take one other shot. This is dominant firms. This is her whole thing. Dominant firms could use their control over these key inputs and could exclude or could discriminate. This is her whole rubric. She's changed the game from, there's actual law breaking going on to this might happen. And then she closed by saying, can we continue to be the home of world-leading technology without accepting race to the bottom business models and monopolistic controls that lock out higher quality products or the next big idea? Yes, if we make the right policy choices, well, I have zero confidence that lean account in the U.S. government are gonna make the right policy choices. And all you gotta do is look at crypto to see how the failure of their guidance and their ability to regulate when it comes to technology. It's not like generative AI was actually invented last year and last fall. It's been around for years. And where has the government been? Yeah. Well, this is again, this is again, this should be a masterclass of the law makers, because they're idiots, most of them. And I won't say all of them, but most of them are. She makes a comment here. This is to protect everyone, collusion, monopolization, mergers, price discrimination, unfair methods of competition. She's got to understand that what the game is, like I said before, and that's kind of, she's just not educated in that area. A platform is an enabling platform that looks like collusion. You can say, you know, I mean, we just covered Palo Alto networks was building this killer security platform, CrowdStrike, among others. You know, you got Amazon, so Hyperscale is a platform that allows companies to create economic value on top of it. That could look like collusion. Microsoft, I mean, if that's the definition, they're collusion. Platforms create opportunities. That's the game. So I think we're in this global platform, they're just gonna, they don't understand what they're doing. She is correct to table some of the issues, you know? I agree. Fraud is legit. We've been covering that on siliconangle.com is rampant. Fake website phishing emails are like unbelievably great with AI. So, you know, you can get an email from someone that looks exactly perfectly personalized for you. Again, for every good side, there's a bad side. So I agree that this should be a look at both sides. Absolutely. But this is not, you don't throw the political agenda. It's totally biased. She's tipping. She's not a neutral third party. She is a biased position that's gonna try to bring down tech innovation because she doesn't understand platforms. And using the UK competition market authority to do your dirty work. Yeah. She's just, to me, it's wrong. It's, she's not transparent. The deck is stacked and she's gotta go. All right. So we're done with the rant section. It's gonna be a downer segment there, but it's kind of important. Let's end on a positive note. I think, Dave, I wanna get your thoughts on it. The entrepreneurship cycle will be probably 10x faster and 100x more powerful in terms of impact over the next decade or two. If you look at the entrepreneurial equation, the ability to go with next gen cloud and data, I think you're gonna see startups come out of nowhere and hit the scene with a blaze of glory and a blaze of flame out as fast as fast than we've ever seen before. We'll see more startups. We're gonna see more startups that are integrating with other startups more on the biz dev side connecting with APIs and cloud. And you're gonna see more economic value. It's gonna be a really great market. I think once we get through this nuclear winter, you know, like I said, the haves and have nots right now, the ones that aren't on the right side of history and the ones that are, but you will see, I guarantee you'll see entrepreneurs come out of the woodwork go. I can use an advantage with data, leveraging super computing, super compute and super cloud and build an app that could literally get herds of millions of users overnight. And these new use cases are gonna emerge and that's the best part about the innovation of America is the longstanding commitment to fostering, you know, a laboratory of innovation. Silicon Valley has done that. New York, Austin, you got Chicago, you got Seattle, you got Boston, you got global markets developing, great entrepreneurial energy. That to me is what the focus should be on, not regulating it and creating rules for a game that they don't understand, which would end up stalling innovation. Every individual with access to technology is gonna be using AI, taking advantage of AI. And yes, I understand there's bad actors, there's good in this world and there's not good in this world. And my belief is the good will outweigh the bad. Certainly, at least for the near to mid to our lifetime long-term, I can't predict, like there are a lot of really smart people are saying, hey, once we get machine intelligence that is equal to humans, we're in trouble and I can't predict that far out and that may be the case. I think that there's a long way to go before that happens. But so I'm like you, John, I'm very optimistic. I think from a startup standpoint, you're gonna be able to do some amazing things a lot faster with a lot less money. And I'm super excited that you said it in one of these pods that you wish you were 25 again. My next birthday, bring me a cake with a 25 when I play. Oh man, if I was 25 and you had all the energy in the world to crank out some great ideas and just drive forward so much opportunity right now, they really make an impact. Not just for money making. I mean, societal impact is huge. AI for good was kicked around a few years ago. That's got to come back and get reset and remodernize them. I think you're gonna see a lot of this Gen X generation do that. And I tell you, I'm pumped. And I think where it's gonna come from, it's gonna come from open source software. We're gonna be next week, we're gonna be up in Vancouver for the open source summit put on by the Linux Foundation. Global Powerhouse has just done an amazing job promoting open source software and open source software is going on on a continued run of more growth. That creates free building blocks and community. So I think you're gonna see that become a real wild card and I gotta tell you, it's gonna be exciting. We just came back from a more pragmatic event called RSA for security. That security show really kind of highlights the importance of data and AI and these super stacks elements because security is a huge problem. We get ransomware, personal tax, fishing. There are huge, huge challenges. So again, the cube is out there covering it and we're gonna be out there. Yeah, John, when we got May, June, always a busy time for us looking forward to see watching you up next week in Vancouver. We got Dell Tech World coming up and then June is super busy as well. We've got Snowflake Summit. We got HPE Discover. We got, we might be doing another show out of the studio. We got Red Hat Summit as well in Boston in May, same week as Dell Tech World. The cube is looking, angle are growing. Vice media is going under. We've covered that media explosion. We're on the other end. We got a tailwind. So the cloud AI is an area people are seeing action in and super pumped. And if anyone has any suggestions, this is our 10th episode. Our goal was to get 10 under our belt. We hit that milestone. We had our first guest, John Chambers. Any guest we should get? Recommendations, DM us. I got a lot of great DMs last week, Dave. That said they loved the band. They loved the back and forth. They wanted a little more enterprise love. So we got the earnings in the air this time. They love the rant section. So maybe we go outside of Lena Connick and get some good rants going too. I think we hit her up, what, three in a row? Three weeks in a row? Hey, she keeps coming back for more. Well, Elon was completely out of the picture this week. So there was a great Bill Maher interview with Elon Musk. I thought it was a good interview. People thought it was terrible. But, you know, Maher was basically tipping his hat to Elon and Elon was being, you know, you can see how goofy he is. And you say, this guy's running Twitter. So it's interesting. It was interesting to see how the fan base responds to more Elon. But people don't like Elon because of his political belief or what they infer to be his political beliefs, but how can you not respect some of the innovations that he's driving? I mean, it's just, that's interesting. I didn't see that Bill Maher interview, but I guess, you know, that's my take anyway, is people just have instant negative reaction. I think he pinned it on his things. We'll see. Anyway, Dave, great episode. Let's grab it up. Thanks for listening and watching. Again, it's on YouTube. It's on Spotify and Apple iTunes. I'm kind of bugged with Substack. I have to upload manually, Dave. So it'll be all the updates on the Substack will be up soon. Of course, go to thecube.net for cube coverage and siliconangle.com. Going great and see you next time.