The Rise of Sovereign AI and Global AI Innovation in a World of US Protectionism

Frank Meehan [00:00:00]: We just saw the power outage in Spain, right? Oh, yeah. What happens if AI is in the banking sector and it just suddenly collapses or takes over? All right. Hi, I'm Frank N. I'm co founder at FrontierOne, a leading sovereign AI specialist working to bring AI to sovereign sovereign in frontier markets around the world, Asia, Latin America, Middle East, Africa, and also working very closely with Nebius, which is one of the, if not the leading AI cloud infrastructure providers in the world and very excited to be working with both of those companies and opportunities.

Demetrios [00:00:51]: And.

Frank Meehan [00:00:55]: Secondly, how do I take my coffee? I take my coffee, I have three cups in front of me. All of them are Americanos with oat milk usually in them. There's occasionally a cappuccino or a flat white may sneak in to that as well. It's a fairly flexible option.

Demetrios [00:01:18]: We wanted to talk about sovereign AI, so I brought you here. I think you know a thing or two about it. You've been having a lot of conversations. Let's just get into it. Why sovereign AI?

Frank Meehan [00:01:30]: Sure. Well, I think it's. Let's start with the market itself in AI.

Demetrios [00:01:36]: Right.

Frank Meehan [00:01:36]: So I've had a, you know, a long history in AI, right back from being on the board of Siri in 2009 when no one in Silicon Valley cared less about AI. I mean, really didn't. And it was so interesting. And then, you know, when I was working for a major family office out of Hong Kong, Horizon Ventures, and we invested early into DeepMind as well, which was 2011, I think, when we did that, so extremely early. And you know, it was like, how do you pick those opportunities? How do you see those trends happening? And I think over the years, just working globally in so many different markets. I've been in Hong Kong and Singapore and Silicon Valley and London everywhere. You get a feel for things and you get a feel for how things are happening. The way trends are moving and I think the way the market is super interesting.

Frank Meehan [00:02:26]: So you have the US and the US is from an AI perspective, it's all about the hyperscalers, Right. So you have the big guys who are providing the GPU cloud infrastructure, like Nebius for instance, and coreweave. And then of course they're providing that to their major customers are OpenAI, you know, the hyperscalers.

Demetrios [00:02:50]: Right.

Frank Meehan [00:02:50]: And also very, very heavily funded AI startups.

Demetrios [00:02:54]: Right?

Demetrios [00:02:54]: Yeah.

Frank Meehan [00:02:55]: So in the us, customers of all these massive AI data centers and, and, and are effectively the big hyperscalers.

Demetrios [00:03:03]: Right.

Frank Meehan [00:03:04]: Because that's where the money is. Like there's a trillion Dollars that's gone into AI in the us in terms of investments outside of the us, it is a totally different picture, right? In Europe, it's like, how much is that? What, $10 billion, I think last year was invested into AI startups. It's a fraction of the spend, right? So where is the AI compute? Where are the AI opportunities? And you know, I've been looking at this for a couple of years and really, Nvidia, to be honest, picked this early. They saw that Jensen has been extremely, extremely keen on the sovereign AI element because he saw that effectively, if I'm a government and I want to deploy AI into a model, right, I want to put my, my data, my defense data, my finance data, my agriculture data, my health data into a model. That data cannot go to the us, it cannot go to China, it cannot go to my neighbor. It has to be in my country. It has to be localized. And no one was prepared for this.

Frank Meehan [00:04:06]: AWS is not prepared to be a local player. AWS is a regional player. I have a hub, I use aws. It's either I use the German hub or I use the, it's, you know, the, the U.S. u.S. Or the Irish hub. Nothing has been this localized ever. And suddenly localization, and suddenly as well, the, the recent developments in the US around from a political viewpoint have forced countries to go.

Frank Meehan [00:04:33]: We cannot rely on the us we can obviously must be wary of China, we must be wary of all these big players. So we have to look out for ourselves. But what do we do? And that's where it's become extreme national priority to have AI implementations because I have to do it local and I have to put my, my data in there and I have to make it secure, I have to have localized models, localized applications, et cetera. And so sovereign AI has gone from a small thing to suddenly be one of the biggest implementations. I think of AI from a rest of the world market outside of the.

Demetrios [00:05:10]: U.S. yeah, it feels like it is something though that is like you're steering a gigantic cargo ship in a way because it takes so much to stand up these data centers. But I don't know, maybe you're the guy on the inside. How long term are we thinking here? And I know everybody's scrambling, but it's not like you can just turn a data center on from nothing.

Frank Meehan [00:05:37]: Well, yes and no. It depends on the size and depends on what it's going to do. So if I'm going to have. So if I'm like, you know, a market like France or something and I want to have the latest Blackwell chips from Nvidia. So I'm going to need state of the art, liquid cooled, state of the art data center.

Demetrios [00:06:00]: Right.

Frank Meehan [00:06:01]: Then that needs to be from the ground up. There are very few data centers, if any, in Europe who are ready for that. Just a few. Obviously Nebby is being one of them.

Demetrios [00:06:10]: Right.

Frank Meehan [00:06:10]: Who are by far like probably leading in the ability to do that. Two, can I even get hold of those chips?

Demetrios [00:06:17]: Right.

Frank Meehan [00:06:18]: So even if I'm going to build that state of the art data center, can I get hold of them? And again, I'm going to need a partner, you know, like an Nebius or call Weaver, someone to do that, who has Nvidia backing very strong and to.

Demetrios [00:06:32]: Just rack them and to figure out how to plug them all together, all the wires, where do they go, what did the box look like?

Frank Meehan [00:06:40]: And the sophisticated sort of, you know, cloud management, Kubernetes, Docker type management that they have on top for managing all those different chips. Because they don't just have Blackwell, they go all the way down, different types of chips or different usage cases. But what's interesting though is in other markets. So let's say Indonesia. So Indonesia were like, okay, they were very smart. I think they worked with Nebius and they were like, okay, look, we don't need Blackwell chips. Probably H2 hundreds, H1 hundreds are fine.

Demetrios [00:07:08]: Right.

Frank Meehan [00:07:09]: And what they did is they got, they took a, they. So the, the government there said, okay, we're going to work with Hutch Endo, which is the major telecom provider there. We're going to take one of their existing data centers and we're going to repurpose that. So we're going to upgrade that so it can take the chips that it's capable of taking. And Nvidia did that work and we'll get it up and running. We'll put a local model on top of applications in space. And it took them four months to put it up and running.

Demetrios [00:07:37]: Wow.

Frank Meehan [00:07:37]: Right. So I'm looking at an 18 to 2 year period if I'm going to try a massive liquid cooled 100, 200 megawatt facility. But if I'm just putting up a 5 to 10 megawatt facility with like chips that are suitable for the needs that I have. And to be honest, like probably mostly inference, not so much the modeling.

Demetrios [00:08:00]: I was going to say that then.

Frank Meehan [00:08:01]: Yeah, I mean there's really different ways of doing things. So it's yes and no. Depends on the, on the size and scale of what you're Trying to do.

Demetrios [00:08:12]: What are different countries ways of approaching this and how do they differ?

Frank Meehan [00:08:17]: Yeah, so you know, Indonesia and you know, the Philippines or, you know, Morocco, etc. All these markets that we at Frontier 1 are working with. So we're, you know, we're, we're, we're basically what we call sovereign AI specialists, right? So we will work with governments on a local level to figure out what their needs are. And it's very local because you'll have to have a local company. You'll be using local talent. You have to bring in the partners. But still they will have to go through a local company that is, and very close in tune with the government and the corporates on the ground.

Demetrios [00:08:53]: Right?

Frank Meehan [00:08:54]: So, so each, each market says, okay, what are we going to do? We're going to have a 3, the biggest data center in the world, like the 3 gigawatts that India is trying to do. Are we going to do one gigawatt like France? We're going to do 100 megawatts like some of the other markets in Europe. And how are we going to do that? Well, we're going to need to establish a local company. And let's say the government's going to go, Someone's going to have to say, where's that customary coming from? Say the government says, I will take up 25% of the compute, right. The other 75 or 30 or 80% or whatever has to come from somewhere else, right? And that's going to be government. That's going to be corporates and research facilities, et cetera. And then someone's going to have to figure out the financing. So either the government says, I'm going to finance all of it, I'm going to finance part of it, and a partner needs to come in and finance the rest of it.

Frank Meehan [00:09:45]: So for the markets that, you know, I, you know, that we work with at Frontier 1, like as we said, across Asia and Middle east and Latin America, all of these markets, it's, it's maybe a 5, 10, 10 mega facility, right? It's 4,000 to 6, 7,000 GPUs. It's 100, $200 million. You know, that type of thing is very doable in those markets, right? So if you, you put that together for them, you find who the right partner is. You know, we, we really like, we really working with Nebius, for instance, because they're very flexible and international, but like, we find that right partner and who's the. And then there's the other elements, right? So, you know, you introduced us to those great guys in Italy who are looking at sovereign AR and Rails. Right, what's it? Pre.

Demetrios [00:10:35]: Prem. AI.

Frank Meehan [00:10:36]: Prem AI.

Demetrios [00:10:37]: Right.

Frank Meehan [00:10:38]: Great team, right. Spoke to them earlier today. Really smart, right. And they've cottoned on to. Yeah, how do we do this for these markets at the smart way, in the lower, in the lower level. I spoke to another team earlier today who are building edge solutions. So like competing with Nebbyus but for remote market, for remote locations.

Demetrios [00:11:02]: Right.

Frank Meehan [00:11:02]: Where power is low, sun is hot, you know, dust, you know, edge locations, right. So you gotta think, okay, where is this thing gonna be? What's the market? Who are the user cases, who are the customers? And for governments they've got like, it's really complicated to try and figure all that out. And so and you, you go to, you're not gonna go to McKinsey and ey, they're not gonna have a clue either. So you know, you've gotta go to the specialists and you know, so people like us who like no software, no infrastructure, know like how to do the financing, you know that that type of approach is I think the right way to go about it. But we will see.

Demetrios [00:11:41]: Yeah, it feels like the different governments are interested in AI because they have to be. We kind of established why it's not all right to just now all of a sudden keep doing things with the US as such a close partner and folks are thinking, well if we're going to be building out our AI capabilities, we should invest in our own country and our own AI capabilities as opposed to shipping off these different workloads or inference to a far away land. And even a lot of these examples that you're talking about, if it's inference, it's going to be much better the closer it is to you. So it makes a ton of sense that you would want to put up different data centers in different countries so that things can hopefully be faster and more enjoyable for the end user. I'm wondering about how you feel governments will change when they now bring online all of this AI capability.

Frank Meehan [00:12:59]: That's really interesting. I mean regulation and data protection and sovereignty and everything comes into that question for sure. The question is, look, I think governments know that they can be more efficient deploying AI. They're not sure how often, but they know they can.

Demetrios [00:13:20]: Right.

Frank Meehan [00:13:21]: And so let's take a very simple example. If I want to put AI models or whatever or applications into my health system and make it more efficient to do that, I gotta be sure that that health data can go into a local model that's super secure, but at the end of the day I want to be able to provide that access to that data to companies that are going to be in startups that are going to be building stuff on my, to make my health system more efficient. So for instance, you know, there's probably a local player that's working on digitizing health records or regional player. So that player needs to have access to that data, but it needs to be secure.

Demetrios [00:14:05]: Right?

Frank Meehan [00:14:06]: And it needs to be. And the government needs to feel that I can provide that access on a secure basis. Now, it's not like I'm going to do everything that way. I'm going to obviously using OpenAI and I'm doing others, but I need to, I need to be very sure that that data can be locked down and secured. And I also need to have a local vector database, a local cybersecurity platform, a local, local model.

Demetrios [00:14:32]: Right.

Frank Meehan [00:14:34]: That provides all that. I think governments are trying to figure all that out, but effectively what they really, at the very simplistic level want to do is can I provide, can I put AI securely into my, you know, into government and provide access to companies that want to do things in health or agriculture or whatever it is securely and don't feel that my data is going to leak because if it does leak, that's going to be on the front page of the news and it's not going to be good for us.

Demetrios [00:15:05]: It's funny that you mention, hey, governments want to do this but don't necessarily know how. And not having sovereign AI to me feels like 100% a blocker on them iterating and trying things so they can't figure out how until they start iterating. They try and put it in here, put it in there, realize maybe that's not really where the value is. So let's try again. You can't do any of that experimentation until you cover your bases and until you get that sovereign AI up and running, you're kind of sol totally.

Frank Meehan [00:15:44]: And I think that's. That, that's absolutely right. It's, you know, they know that that's a starter package. And let's just look at research, right? Universities and everything that need to do research. Again, that's a lot of government sponsored research. And again, you want to keep that secure.

Demetrios [00:16:02]: Right.

Frank Meehan [00:16:02]: I need to be sure that that data hasn't gone into someone else's model in say, Silicon Valley somewhere.

Demetrios [00:16:10]: Right.

Frank Meehan [00:16:10]: And I want to, you know, have that aspect to it. I don't, I mean, a lot of People have seen sovereign AI is like government control and it's not, it's, it's government, it's, it's trying to figure out how do you like look after the data in the right way. That is very important to you. I mean know your citizens health records are very, very important to you. You cannot let that data go to another nation and neither your defense like data.

Demetrios [00:16:43]: Right.

Frank Meehan [00:16:43]: You're, that's also extremely important to you. You cannot let that leak. So it's, it's tricky but it, it needs to be done. But then at the same time you can't do everything. So you're going to have these great startups and companies that probably local or regional that are going to want to work on the, on the, the system and you have to put that in place for them. You have to have the rails for them to do that. Which is what pre mii. Yeah.

Frank Meehan [00:17:10]: Which is what those guys are doing. Which is really cool. And so I think, you know, it's our job at Frontier 1 is to bring all that together and be the master builder right. In these markets and help them understand all those aspects and also always keep on top of it and make sure best of breed is coming through the door.

Demetrios [00:17:31]: I wanted to just zoom in and almost like push back on the medical data one for a second. The defense one. I think everyone 100% agrees you want to be doing your defense data in your country and that is without a doubt number one priority, the medical one. I often wonder if it's not better to get the more medical data the better because you're going to find these random very long tail type of diseases and you get to see well, where else is that happening? And it could be highly valuable for researchers. I, it's also funny that you mentioned this because I was just listening to the acquired podcast on my chart, which is a company, I think it's epic. The, the company and they created this thing called Cosmos and it anonymizes hundreds of thousands if not millions of data, patient data records and then researchers are able to go and look at that and even doctors are able to go and look at that. So in that regard, I don't know.

Frank Meehan [00:18:47]: Yeah, I mean I will go with that. Say look, it's, it's two aspects to it, right? Anonymized data for research purposes. 100%, no question. But the question is when you know, if I'm going down to the NHS here and they're pulling up my record and they're trying to figure out like what's that's an internal NHS system which is accessing my record. And that absolutely needs AI because it's a disaster of a system, clearly.

Demetrios [00:19:15]: Right.

Frank Meehan [00:19:16]: Because nobody, I mean, they sit there and go, you know, it's like, whatever.

Demetrios [00:19:20]: Right.

Frank Meehan [00:19:21]: That's a different problem.

Demetrios [00:19:22]: Right.

Frank Meehan [00:19:23]: So. So there's two parts to that problem. One is the research on anonymized data. No question. I need as much data as I can. But of course, the other question is the localized NHS equivalent is accessing highly sensitive private data. And that's a different problem. So there are two problems to that.

Demetrios [00:19:42]: I see what you're saying. So it's more of this. When the inference side of things or the execution of the workloads are happening, you don't want them getting even if it is fully anonymized, it's better to keep that close to home.

Frank Meehan [00:20:01]: Yeah. And especially if I'm going to be modernizing, let's say I'm modernizing the NHS here in the uk.

Demetrios [00:20:07]: Right.

Demetrios [00:20:08]: God can only hope, huh?

Frank Meehan [00:20:09]: Yeah, but. Exactly.

Demetrios [00:20:10]: Right.

Frank Meehan [00:20:10]: But that or anywhere else.

Demetrios [00:20:12]: Right.

Frank Meehan [00:20:12]: Those systems are old.

Demetrios [00:20:13]: Right.

Frank Meehan [00:20:14]: They're 10, 20 years old in all of these markets. Right. At least. And they radically need updating. And that's going to be a radically different way of doing things. And they're going to be doing modeling on that huge amount of data that they have that is not anonymized for that purpose because they need to also use that internally. So it's a very different problem. So that's the one I'm going to be referring to more here.

Frank Meehan [00:20:40]: From a sovereign AI perspective, anonymizing data from a research perspective, that's a different problem. And yeah, I mean, give the research as much data as they can, for sure. But what do I do? So, for instance, can the AI, let's say I go in and get a body scan here at the nhs.

Demetrios [00:21:00]: Right.

Frank Meehan [00:21:01]: And they use that data and they model that data against others and they come back and they tell me X about me.

Demetrios [00:21:08]: Right.

Frank Meehan [00:21:09]: That is very sensitive data that needs to be protected.

Demetrios [00:21:13]: So.

Frank Meehan [00:21:14]: But the people that are going to be developing that application need to access my data. So, you know, that's. That's a different problem in a way.

Demetrios [00:21:25]: Yeah. Also, you have the EPIC company in the us, but every country has their own EPIC or a few different epics that want to be creating these ecosystems so that folks can be pushing forward in the progress area. And I think I fall victim of this a little bit because I am American of thinking that there is the US and then everybody else and that's kind of, we don't think about that, obviously. Living outside of the US for so long, I recognize there's this great big world with a whole lot of people out there, and it needs these type of things too, these initiatives and all this stuff. And when I was in the, when I was in the Zurich Gen AI conference a few weeks ago, one thing that was top of mind for them in a few different talks that I saw was, hey, all this AI stuff is happening. How is Switzerland being part of it? And of course, Switzerland is traditionally very neutral. So for them, sovereign AI felt like a natural next step or a natural path to take. And it, it also helps that they can say, yeah, we're going to go invest in this project.

Demetrios [00:22:50]: I wonder if you see different countries or scenarios where maybe they want to invest in it, but like, they don't understand or is there just budget right now? Everyone understands that this is as important as possible.

Frank Meehan [00:23:07]: I think it's a great question. I think taking a step back again, like, what's, what Trump has done has, is, is end globalization, right? And not just in trade, but in how things are built, right. Traditionally, a company would have been a country doing completely fine using a US company to do whatever they did, right?

Demetrios [00:23:30]: No more, right.

Frank Meehan [00:23:31]: He has turned everything to localization instead and in some ways incredibly cut the foreign business development opportunities for a lot of US Companies, which is incredible, right? And so, but on the other side, that means that a huge opportunity has opened up for local regional companies around the world. It's just transformed the landscape incredibly, right? So now Europe is going to go, I need as few US companies in my sovereign AI or banking facility as possible, right? And so I need local and the local. Suddenly these local, like it's, it's so interesting. I mean, you know, suddenly they're going to be, I need local, local companies for this and I need local companies for that and, and all elements of the stack. There are some banks in Germany have said, like, I cannot have a US company in the stack, right? I can't even have GitHub. So like, that's, so that's huge, tremendous opportunity for companies around the world to suddenly become taking advantage of that. And that will start primarily in the finance and government sectors because both the finance, so finance, defense and government will suddenly go, we cannot, we can't use US Companies. We may not even be able to use a UK company.

Frank Meehan [00:24:57]: We have to use a European company. And UK is like, yeah, we need to do this. And Saudi is, well, we're going to need to do Saudi. I mean, it's just tremendously opening up the opportunity globally for local talent and local development of core AI functions. Just from the geopolitics of everything that's happened, I mean, Trump has literally suddenly woken up the world who was like, pretty much reliant on US Companies in US Software to go, wow, well, wow, what do we do? So, I mean, it's kind of at that level now, right? Everyone is a bit. Just at that conference that you were at, everyone's like, well, what do we do? And so, yeah, I mean, that's where we come in and go, okay, let's help you through this. From a local level, there's going to be tremendous opportunities. I think AI is truly going global thanks in large part to Trump.

Demetrios [00:25:56]: So, yeah, he's great for your business. You've got to send him a thank you card or something.

Frank Meehan [00:26:01]: 100%. I mean, there are a lot of people that are in the, you know, technology space who are going to be doing that now because it's the same thing. Like, if I'm, you know, let's say I'm a. It just applies to so many things. If I'm a. Again, if I'm a country and I'm, let's say there's this kind of rumor that the F35s that if I buy an F35, the US can switch them off. I mean, that's super interesting right now.

Demetrios [00:26:32]: Now you gotta look for a new provider.

Frank Meehan [00:26:33]: It's got a Remus, whatever. But it's entirely possible that the US could switch an F35 off.

Demetrios [00:26:41]: Right.

Frank Meehan [00:26:42]: So I've just spent a billion dollars buying an F35 and you can switch it off with a kill switch. I mean, I, you know, like, that's really interesting.

Demetrios [00:26:50]: Right?

Frank Meehan [00:26:50]: And he has, as we just saw from the Canadian election, like, completely transformed countries. Canada is now more nationalistic and more local than ever, and that is happening in many markets. And it's going to open up a lot of opportunities for AI developers and stuff all around the world.

Demetrios [00:27:10]: Yeah. First off, let's just have a moment of silence for those engineers that have to work at the banks that don't allow GitHub, because poor folks out there.

Frank Meehan [00:27:21]: You know, they are going to be.

Demetrios [00:27:22]: Going through hell and back. But the other thing, just think about.

Frank Meehan [00:27:26]: What they've been having to use instead.

Demetrios [00:27:28]: Yeah, that's what I'm thinking about. It's like, oh, man, that's going to be rough. But the, the other piece was around this idea of we've seen the US weaponize the dollar and like you're saying there's this jump or there's a step where they can weaponize the weapons that they've sold to other countries. So if you get on their bad side and they say ah, potentially they just turn off their jets that they sold to you, whether or not that's true or not, whatever, it doesn't matter because the thought has been planted into different governments and politicians minds that they are making an effort to move away from the US as the supplier. But then if you extrapolate that forward, you can think about it in this frame of well, what's going to stop them from weaponizing AI? And maybe today with the use cases that we have, it isn't as strong of an argument. But potentially in five years or sooner or later, maybe it's 10 years, that could be a whole different environment that we are living in. If we are as reliant on the US's AI ecosystem as we are on the dollar, then if for some reason you get on Trump 3 or whoever the next president is on their bad side and they say well, we're just going to turn off your AI, that is another way that it can be weaponized. So I fully see the other side of this story where it's like yeah, we gotta make sure that if AI is as transformational of a technology as most folks are saying it's going to be.

Demetrios [00:29:19]: Whether or not it is, I, it's debatable, a hundred percent. I think there's a lot of early signals that it's going to be, but even if it isn't, it's better to have it and not need it than to need it and not have it per se. So you don't want to the US to have that capability of just like being able to shut off your AI.

Frank Meehan [00:29:41]: Yeah, look, I mean, I think the us at the end of the day is, is just hyper focused on itself right now and, and that's fine, right? I mean that's, that's really fine. I do think that the Trump administration will, will gradually, you know, start to, and is starting to realize that you know, maybe this is, you know, things will change.

Demetrios [00:30:02]: Right.

Frank Meehan [00:30:02]: So I do think it's just hyper focused on itself and just making sure that the US has the right resources and the right chips and everything that are available to it and, and like it's going to go like we need the best chips, we need a lot of them and we need to, we need to conserve them.

Demetrios [00:30:16]: Right.

Frank Meehan [00:30:16]: And the US is Right, completely. I think that, and that's like a lot of markets are doing that.

Demetrios [00:30:24]: Right.

Frank Meehan [00:30:24]: China does exactly the same thing. So the US is allowed to do that as well. I think at the end of the day, you break the US out from the rest of the world. And that, as we said, like globalization, for whatever reason is coming the, a bit to an end in a lot of ways. And localization is coming back again. And that means that by having a local, especially for something as important as AI, having a local deployment that I can rely on is important.

Demetrios [00:30:55]: Right.

Frank Meehan [00:30:56]: And that I, you're absolutely right. You know, can it, you know, something, maybe they do switch it off.

Demetrios [00:31:00]: Right.

Frank Meehan [00:31:01]: And that, that could have a huge impact because I have AI throughout my infrastructure.

Demetrios [00:31:05]: Right.

Frank Meehan [00:31:05]: I mean, we just saw the, the power outage in Spain, right?

Demetrios [00:31:09]: Oh yeah.

Frank Meehan [00:31:09]: What happens if AI is in the banking sector and it just suddenly, you know, you know, collapses or takes over or, you know, so that's wild.

Demetrios [00:31:19]: I saw stuff that folks were saying they couldn't get cash out because There was no ATMs there. They couldn't pay for anything because all of the credit card machines weren't working. And so effectively it was like, well, I guess I'm not going to do anything for a few hours.

Frank Meehan [00:31:35]: Yeah, that's wild, right? I mean, now that's an energy problem which, you know, comes back online. But again, what happens if I've got AI, you know, heavily into my energy systems and everything, so it really has to be protected. And I, I, I do think that, yes, you can make your own mistake nationally and you can fix it, but if you've got another actor considerably embedded into your AI local infrastructure, you need to be careful of that.

Demetrios [00:32:04]: Right.

Frank Meehan [00:32:05]: And that, you know, the US isn't going to allow China to run their AI infrastructure and neither should anyone allow anyone else to write. I mean, like, that's the localization of it when it comes to the government side, which is so important. I think, especially if you look at, you know, energy and finance and where we could be in five to 10 years, where AI systems are AI agents and know, potentially, you know, general intelligence is, is, is running these things and we don't even have much control over them. And we're all gone to Skynet, which is not, doesn't seem that far away now. You know, there is that potential of, you know, that, that famous scene in T2, right, when you know, the guys are switching on Skynet without having any idea what it can possibly do, that I think is going to be a lot of Governments and so like you've got to be able to how you're going to implement that and control that at least at a national level is important.

Demetrios [00:33:02]: Yeah.

Demetrios [00:33:03]: What are some of the other key players in this? I mean you mentioned how you have to have the government's bought in, you have to have financers. You almost want to be looking at the different tools that you can offer in these data centers too because it's not just the straight hardware, you've got the software. Is there anybody else that I am missing?

Frank Meehan [00:33:25]: So I think you know, you look at the stack. So I mean like firstly I've got to go, let's say I'm implementing whether it's like I'm repurposing an existing data center, but let's say I'm building one from scratch. I got to find the right power, I got to find the right location and also very importantly before I build, I got to have the right customers. So I need to know before I'm committing a billion dollars or even $100 million of GPUs, who is the customer for this?

Demetrios [00:33:51]: Right.

Frank Meehan [00:33:52]: So the government or someone has to say I'm going to take the off take for that. Now in France, Macron twisted everybody's arms and said okay, you all these corporates are going to come together and we're going to all contribute and say basically put a billion dollars worth of compute off take spend on the table. I think that's very smart and that's a really good way of doing it. So I do think the other players are going to be big corporates in local markets. He will work in tandem with the government to do that. For instance in Argentina, you know, the government there has been really, really strong that they want to make Argentina a local AI center and the corporates are being brought in to figure out the usage of it because at the end of the day who's going to use it, how's it going to go?

Demetrios [00:34:40]: Right.

Frank Meehan [00:34:40]: So I do think that's the big, another big player is the corporates themselves and figuring out how they will interact with systems locally. And the other players are going to be local players on the ground who can build local applications.

Demetrios [00:34:58]: Great.

Frank Meehan [00:34:59]: A lot of people I'm sure in the mlops community. It's really interesting. I mean you think where I think is so interesting for a lot of markets instead of just being a little outsourced team of some big US company, the, the opportunity is to, is to build local and build something and then potentially regionally and, and, and keep going We've seen in the Middle east there's a, lots of great regional players who do things in the Middle east, you know, because of different requirements, not least language, but, you know, so how does that, how does that come together? It's really interesting.

Demetrios [00:35:38]: Do you feel like that leads to a world of more quote unquote lifestyle businesses where you get smaller teams, they have smaller markets, but it's less venture backed or less fundable because they're not going to go and conquer the world and there's not that infinite tam.

Frank Meehan [00:35:59]: It's a great question. I mean we've obviously been talking about that opportunity and I think it, it certainly enables the tools that are available for everybody now. I, you know, we can, we can spin things up faster with smaller teams. Yeah. And we can use tools to, to run things. Right. We can have agents running customer service and we don't need to pay $10,000 to some team, some group of people who are going to do customer outreach for us. We could probably use agents much more effectively now.

Demetrios [00:36:27]: Yeah.

Frank Meehan [00:36:28]: So I think people can spin things up and prototype things much, much faster. They can come together in little teams to do things locally much faster and solve local problems. And those local problems can go regional.

Demetrios [00:36:43]: Right.

Frank Meehan [00:36:43]: I mean, I'm a partner in a great fund that is out of Oman in the uae and a couple of those companies, even somewhere as, as small and as tight as Oman, have gone regional just because they solved a particular problem that just works regionally.

Demetrios [00:37:02]: Right.

Frank Meehan [00:37:03]: And gradually scale it out and now they can do it faster. And one of those is in the health sector. And so yeah, I really think that's going to be the case. I think it's a change. Right. You know, instead of as we talked about beforehand, like you know, the, the model, the old venture model of giving $5 million or $10 million to a company and saying, hey, let you know, you know, nine months, 12 months, produce something and that's MVPs are great, you know, just limited functionality. Just get it out there. Like that doesn't work anymore.

Frank Meehan [00:37:39]: Like, you know, someone else can prototype something so fast with a lot of functionality and a lot of features and test it and roll it out faster. And I think that's really exciting and I think that's where the power of like what the mlops community has is really powerful. Like that's a group of very great people who are in the AI and ML space and who are thinking about these things and prototyping these things and they're not on LinkedIn. And if they Are they're not going to be really sharing their life experiences on LinkedIn.

Demetrios [00:38:12]: Right.

Frank Meehan [00:38:13]: So they're doing that in a community like MLOps. And so how does that come together and build cool new things and prototypes, I think is the, the next kind of logical step for communities like, like.

Demetrios [00:38:27]: You guys and I know that we had talked about too, these traditional ways going back to that idea of hey, I've got to go now and raise a bunch of money, go all in on my idea and spend the next nine months launching it, bringing it to market, etc. Etc. That has been a bit turned on its head where you can, in a way you can have your job at your big fancy company making all that cash, but still you can be moonlighting on this project which it has the potential of like the iteration speed I think is what is different because you've always been able to moonlight, you've always been able to have your com, you working at your big company. You are now moonlighting for something and then you bring it to market or you put it out there. But the ability to have that feedback loop be tightened is the coolest thing ever. And we're seeing it so much with a lot of these bottoms up companies like the Lovables and the Cursors where they have such a product led growth motion that they can get to a gigantic scale with the small teams because they don't have any sales and marketing teams.

Frank Meehan [00:39:50]: Spot on. I mean look at Lovable, right? Another Swedish, great Swedish know like Spotify, right. You know when Spotify started, you know, like in Sweden, right? You're going to take on the music market from Sweden and yes, they did completely right. And Lovable's done the same thing like you know, because of the ability to scale things faster and smarter and the hyper efficiency focus of Swedes is great. You know, I used to live in Sweden and it's amazing. I don't know how they do it. You know, six weeks off in the year went from sort of eight in the morning till four in the evening, four in the afternoon, but still like Ikea.

Demetrios [00:40:31]: Yeah.

Frank Meehan [00:40:31]: Mobile, Spotify, everything.

Demetrios [00:40:33]: Right.

Frank Meehan [00:40:33]: But I think the point is, is that you're absolutely right, right, Is that companies can, can prototype and spin things faster and I can do like, and you know, I, I would be, you know, if someone's at a big company and they were like a little cog in the dev team and it's really kind of bit boring and they do the same thing over and over again with a little bit more next, you know, In a couple of months, you're going to be looking at other things to do. But those other things to do now are infinitely more interesting and infinitely more scalable and faster to prototype and get out there. But also, I don't need to leave the day job until I know it's working. And then suddenly it's working and then, oh, wow, okay, maybe I'm going to make a jump now. And same as a VC, a VC is going to be like, I think VCs need to change their idea of what it is.

Demetrios [00:41:21]: Right.

Frank Meehan [00:41:21]: I don't need to have 20, 30 people all doing something before I invest in them. I need to have like five very smart people who've launched something. And I don't care how they got it out the door. I don't really care.

Demetrios [00:41:35]: Right.

Frank Meehan [00:41:36]: I just know they got it out the door and it's really cool and it's scaling and I'm going to put those. I'm going to invest into those five people. I don't care where they are or where they are, but I'm going to invest into them. That's a different story. And I think VCs have to change the way they're approaching the market.

Demetrios [00:41:52]: Now, going back to the sovereign AI for a second, because one lingering question that I have is how you see regions cooperating. Specifically, we have the eu. It is very friendly in a way. But you have been mentioning, hey, France is doing this on their own. They're building their sovereign AI system very much for them. I know that Germany has talked about, actually, I think Germany's doing something cool where they're taking old factories and they're trying to spin up data centers with them in the old factories, because the old factories are now defunct. But they're like, well, the structure's still there. It's actually a big space.

Demetrios [00:42:37]: You know, it'd be perfect here. Whole lot of GPUs. How do you see the interoperability of the countries? Maybe we'll start with Europe, but then go towards, like you were saying in the Middle east, there's these companies that will spin up on top of the structure and it starts locally, then goes regionally. Do you feel like these joint partnerships and the data centers are going to be only a specific country thing, or is it going to be like that more regional thing? And how does that play? How does that work?

Frank Meehan [00:43:09]: Yeah, I think it's a great question. I mean, we don't know yet, but we can surmise that, let's say France builds that $1 billion facility or even a $5 billion facility. I do not think that enough customers are in France for that type of facility. There's, I mean, what I mean is there's not enough usage in France for that facility. And I think France will then go, hey, let's make this available to the rest of Europe and maybe just carve out some racks. And we're going to need to give the ability that like some hyper localization on some of these racks. But yes, I do think so. And then is there a EU sovereign AI? Of course the EU is going to do it.

Demetrios [00:43:58]: I mean that's, you know, the most regulated sovereign AI.

Frank Meehan [00:44:02]: There's some guys in Brussels dreaming about this, this right now. And is that going to work? Well, yeah, let's see another question. But they're certainly going to try and they're going to have some sort of European AI, but nobody really even knows what the definition of that is, you know, and, and the problem of course in the EU is it's so over regulated that anytime I try to launch something I've got to check so many hoops that by the time I launch it it's out of date. So I think there's that vast problem of regulation needs to get solved in the EU first as well. I think other markets have an opportunity to not be so hyper regulated and cooperate on a regional basis without the huge regulations that the EU throws in. Things that just stifle every innovation. You know, I think you've got a clear opinion about that. But yeah, so I think it's, it's different, right? Markets are going to be different.

Frank Meehan [00:45:04]: Like how does Latam collaborate? Africa.

Demetrios [00:45:08]: Right.

Frank Meehan [00:45:09]: Asia, like there will be collaboration for sure.

Demetrios [00:45:13]: Yeah, I like that idea of there's going to be excess demand so it's going to get sold and it feels like potentially there's going to be that in these Middle Eastern countries too. And so it's interesting to think, wow, I imagine a world where even if you're not fully bought in, maybe you're not doing something so sensitive that you need the sovereign AI. It's just me, app developer in the EU and I go on the streets looking for some GPU capacity. If there are a lot of different data centers that have been built up around me, that's just going to make my cost go way down.

Frank Meehan [00:46:00]: Yeah, that's right, definitely. And I think that's, that's going to happen. And everything of course at the end of the day comes down to power. So you know, let's take something like Bhutan, Bhutan suddenly has Become incredibly interesting because it's got 30 gigawatts of potential hydropower of which it's only used about 2.

Demetrios [00:46:20]: Right.

Frank Meehan [00:46:21]: Is so much hydropower. So does Bhutan. If they put a little. A major data center in the south of Bhutan where they're building the new mindfulness city, we could have a mindfulness data center down there. But I mean, the point is their power is so cheap and so I mean, they could be a regional player just from the source of power they have.

Demetrios [00:46:45]: Right.

Frank Meehan [00:46:46]: So you know, El Salvador is looking at building Bitcoin city in a volcano.

Demetrios [00:46:52]: Right.

Frank Meehan [00:46:53]: Using geothermal.

Demetrios [00:46:54]: Right.

Frank Meehan [00:46:54]: Hopefully extinct volcano. And now maybe Bitcoin city is, you know, I don't know. Well, today it's probably pretty good idea. Last week it probably wasn't. So who knows.

Demetrios [00:47:05]: Right.

Frank Meehan [00:47:06]: But there's a little bit more, you know, surety around AI. Right. So they're looking at thinking, okay, should we build a data center there?

Demetrios [00:47:14]: Right.

Frank Meehan [00:47:16]: And could that be regional? Like, you know, there's a lot of thinking that goes into it.

Demetrios [00:47:22]: Yeah.

Frank Meehan [00:47:23]: But I do think, yes, there will be a degree of localization and regional support. A lot of driven by which country has the lowest cost power.

Demetrios [00:47:31]: And it feels like from what I've understood, you have to have the governments that sponsor this in a way. Otherwise you're to a certain degree, because.

Frank Meehan [00:47:42]: I, as a hyperscaler or whatever or a BlackRock or anything, I'm not going to put a billion dollars into building an AI data center if I don't know who's going to be using it and who's going to pay for the usage. So yes, to a certain degree the government does need to put in like maybe they say I take 20 or 30% of the offtake and guarantee it. Which means that, yeah, okay, now I feel more comfortable about financing this.

Demetrios [00:48:07]: Is there anything else that you wanted to talk about that we haven't hit on?

Frank Meehan [00:48:10]: Man, I could talk about things for. I could talk about anything really. I mean, like, you know, whatever, really to honest with issues. But I think today's topic has been exactly the key thing, which is really for me going back to it again. AI is truly going global. We saw that with Deep Seq, how that came out of nowhere and you know, and surprised the Chinese government as much as surprised anybody else.

Demetrios [00:48:36]: Right.

Frank Meehan [00:48:36]: And where did that come from? It came from a hedge fund which was looking to do better modeling and decided to build this thing and. Wow.

Demetrios [00:48:44]: Right.

Frank Meehan [00:48:44]: So I think we're going to see a lot of things that are going to expand AI globally And the opportunities are going to be tremendous. And we're just seeing the start of that.

Demetrios [00:48:58]: You know what I wanted to talk to you about for a second is take me back to the Siri days pre Apple acquisition. What and how did you even think like was give me the scenario. Because I think it's a fascinating story on how Siri got rocking and rolling. It was a bunch of researchers, right?

Frank Meehan [00:49:19]: It's a bunch of researchers out of SRI International. Um, and it was really like, you know, I have to say, you know, my, my boss at Horizon Ventures, she's incredibly, incredibly smart and she saw the opportunity. She was thinking about data and AI way before anybody else was.

Demetrios [00:49:39]: Wow.

Frank Meehan [00:49:40]: And it was together with. And this is out of Hong Kong, right? And, and she's so smart. And the others that were in it was Menlo and Morgenthaler and like kind of older VCs from the 90s that knew deep tech, right? And we're like, deep tech is still always going to be important. In those days, everything is just about social, social network scaling. That's where all the money was going. But you know, Manilow, Morgan Thaler and us were like thinking a lot longer term about what are the impacts of deep tech. And this is like deep tech wasn't even a thing. Like, you know, people aren't really talking about it, right? And these guys led by Adam Chaya, who's super, super smart, just great guy.

Frank Meehan [00:50:26]: I'd been working for 10 years in Sri on AI, right? And the opportunity came to do something and spin it out. And look, it was so under the radar. We launched in February 2010 on the app Store when the App Store was very small and it did sort of help you with reservations and cinema bookings and things. And there was one also the other guys that were very forward sighting was Verizon. There was a guy called Ed Ruth at Verizon who was very forward seeking and had done a deal between Verizon and the Siri team so that Siri would launch their first.

Demetrios [00:51:05]: Wow.

Frank Meehan [00:51:06]: And it was so interesting. And so basically what happened then was it launched in the store and then Steve Jobs saw it and about two days later the founders got a call which they didn't believe, but obviously, you know, it turned out to be Steve Jobs. And they went into the Jobs reality distortion field completely and it was great. Like four weeks later it was sold 200 million. And that was fast and it was crazy. And the amazing team, unfortunately Steve got sick and Siri got lost internally for a while. But the team then went on and built amazing things and for Samsung and then Airbnb. And it's just, I mean they've just an incredible team that have continually innovated in AI, constantly with new companies.

Frank Meehan [00:51:55]: But yeah, back in those days, I mean, Steve was, you know, his jobs was so far ahead of the thing. He saw it. But I have to say, you know, my boss at Horizons and those guys at Menlo and Morgan that are also saw it.

Demetrios [00:52:08]: Right.

Frank Meehan [00:52:08]: You know, they saw it way above it. People were just, they were just had that long term deep tech focus. Yeah.

Demetrios [00:52:15]: And.

Frank Meehan [00:52:15]: And none of them were bankers.

Demetrios [00:52:17]: Right.

Frank Meehan [00:52:17]: They weren't finance guys. So they saw it from a technology perspective, not from a finance perspective.

Demetrios [00:52:23]: And no one ROI is not a question.

Frank Meehan [00:52:26]: Or DeepMind, when they pitched were like, we have no idea where the revenue's coming from. None. And in those days that was novel.

Demetrios [00:52:33]: Right?

Frank Meehan [00:52:34]: So no VCs doing it. Like hardly any VC would do it. Like, no VC was investing in DeepMind in Europe in the early days. Right. It was like, well, where's the, where's the revenues? Where's the. We don't know, like we're just going to build the best AI engine on the planet. Well, why would you do that? I mean, like, you know, so that kind of question like comes down to it. And I think, you know, again coming back to it, I think there is again a huge shakeout going on where a lot of investors work very traditionally.

Demetrios [00:53:08]: Right.

Frank Meehan [00:53:08]: And there's a new way of things going to happen where people are going to spin things up really quick and jumping on that really quick is going to be the key. So yeah, watch this space.