How will we fund the global AI revolution? >> All the rules are being rewritten about [music] how you fund growth because we just need all the capital we can get. What is the main thing? It is AI. Where does the next Nvidia-style growth come from? >> The compute has gotten so expensive. >> They are going to dedicate [music] massive amounts of capital to this space. I’m the old-fashioned stock exchange. I think our common challenge will be to to make sure that you know we find as many ways as possible that we match the capital with the opportunities. The amount of capital going into the sector way outstrips the venture funds. That trend is now drawing in a huge amount of money, which is why we’re talking about it on this stage in Saudi Arabia. The the untapped but mobile capital is here in this room, and if it jumps on the opportunity, the it’s like an opportunity I’ve never seen before. Now that’s a moon shot, ladies and gentlemen.
[00:01:07] All right, welcome everybody to our AI mini summit brought to you by Link Exponential Ventures. It’s a pleasure to have you. We’re going to be having a series of 30-minute conversations that look at AI investing where the next trillion-dollar companies are coming from. We’ll be having a session of our moon shot summit. And I’d like to open with our first session, which is how will we fund the global AI revolution? Uh to enable this conversation, it’s a pleasure to bring on stage three leaders in this field. David Blundin, David is my business partner. He’s a serial entrepreneur. He is the managing partner of Link Exponential Ventures with 23 startups under his belt, a long track record of a
[00:02:01] 44% IRR, a little over a billion dollars AUM, based on the campus of MIT and Harvard. David Blundin, please come on up. Take a seat here. Thank you, David. >> [applause] >> Next up is Bonnie Chan, CEO of the Hong Kong Exchange and Clearing 8 HKEX since March of 2024, bringing over 30 years of global capital markets legal and listing transformation experiences. Bonnie, please join us. And finally on our panel this morning is Anjan Mitra, partner at Andreessen Horowitz, a16z, investing in frontier AI, open source uh the man who’s backed Anthropic on the board of Mistral. Please welcome to the stage Anjan Mitra. Hey, Anjan.
[00:03:03] So, how will we fund Take a load off, Peter. >> AI revolution, guys? So, I mean, when I think about it, uh we are seeing today, at least in the United States, a billion dollars deployed per day into AI. The expectations are we’re going to see that growing to three billion dollars a day by 2030, and I expect it’s going to blow through that. Uh in fact, I’m seeing capital flowing to the exclusion of a lot of other things. Uh let’s open with opening thoughts around that. Anjan, I mean, you’re at one of the largest venture funds on the planet. Uh what percentage of a16z is flowing towards AI? Uh and what are your what are your thoughts about about the capital availability to fund this, you know, infrastructure, you know, what I what we call on the
[00:04:00] moon shots podcast, tiling the earth in compute? >> Right. Uh How much how much capital is flowing into AI? Basically, all of it, and it’s still not enough. Um because, you know, what what the firm was founded to be a verticalized firm. We have an infrastructure fund, an applications fund, a health care fund, and all of those are now AI funds, right? Because AI is a cross-stack thing, whether you’re you’re working with teams that it we’re training foundation models or building applications. The I I don’t think anybody is not an AI investor anymore. Um on the other hand, what’s also insatiable is the is the need for these for AI businesses, especially ones that generate tokens, um that leverage the latest generation in reasoning models, which generate 10 times more tokens than traditional, you know, gen AI models before reasoning. The the the we’re living through like Jevons paradox every day where, no matter how much infrastructure buildout we do, no matter how many um you know, if algorithm algorithmic efficiencies they are, we somehow just need more compute, more infrastructure
[00:05:00] to serve the state of the art demand in text, code, image, video. It it’s just sort of this insatiable explosion of use cases, and I just don’t think we’ve figured out how to change the traditional venture capital stack to fund all this growth, and that’s why you’re seeing you know, we try to fund entrepreneurs as much as we can, but then we got to pull in all the friends we can, whether that’s Nvidia as a strategic who who invests on the cap tables directly alongside us. It might be a data center provider. There’s just, you know, whether it’s Satya doing a billion-dollar investment into micro into OpenAI as a nonprofit 4 years ago, or it’s Amazon and Google investing in Anthropic. There there’s just all the rules are being rewritten about how you fund growth because we just need all the capital we can get. Yeah. Bonnie, I I’m when I see an offering being made by Elon for xAI or by Anthropic or by OpenAI, instantly it’s filled. I mean, people are fighting to get into these deals. And no one’s asking is the valuation is
[00:06:01] the deal going to make sense. They’re just throwing capital at this. How are you seeing it from your perspective? Well, first of all, I do agree with the comment that Anjan made, which is the insatiable demand, right? I mean, just everyone wants to pour money into it. But I must say, Peter, it’s very interesting how you put together this panel. Because, you know, as I see it, I’m sort of, you know, sticking stuck in the middle of these two gentlemen. You represent the private side, shall we say, the VCPPE community. I’m the old-fashioned stock exchange. I do public offering. I do IPOs. Um and and so, you know, how are we going to fund it? I think there are many different ways, but uh suffice to say, you know, given that, you know, Hong Kong Stock Exchange, obviously we’re in Asia. And I would say, um given the demographics, there is an emergence of a lot of um uh well, a big population of retail investors. We tend to now call them
[00:07:00] protel investors with technology. Now everyone have their own trading theories and strategies and whatnot, and they can execute, you know, rather in a rather sophisticated manner. So, I must say that, you know, from my vantage point, I still think whatever ways is available which can bring as many different pockets of demand, right, from different investors at all corners of the world, will probably, you know, be a good way to support the development of AI on the one hand and really quest that insatiable demand on the other hand. So, um to put things in context, we’ve done quite well this year in the IPO space. In fact, Hong Kong is now number one on the global IPO league table this year. Uh we have we have 300 300 deals in the pipeline waiting to get done. We have already done about 80 year-to-date, and I would say of um the 80 which has been completed and the 300 which is still waiting in line,
[00:08:01] about probably half of it has something to do with AI. Now, there are different manifestation, but I would say especially with the companies in the Chinese mainland, these days if you are not already doing something with AI or being, you know, at the very center of the AI development, you probably quite unable to compete and be successful in your business. So, this is sort of my answer to your question. I I think really um just given how much capital is needed to support the growth, whether it’s private, whether it’s public, whether it’s credit, whether it’s equity, does not matter. Uh I think our common challenge will be to to make sure that, you know, we find as many ways as possible that we match the capital with the opportunities. David, uh at Link, you’re seeing and investing in companies as the first check. Yep. Uh companies born out of MIT at out of the,
[00:09:02] you know, CSAIL computer science AI lab and out of Harvard. Uh what are you seeing as the growth of companies going into AI that is feeding the pipeline at the early stage? I’ll tell you, there’s there’s a reason Bonnie’s on this panel with and sandwiched between the startup guys because the amount of capital required coming into these companies, like you said, three billion dollars a day coming up. US venture is 200 billion a year. So, it’s not even close. You know, five times more money needs to come from somewhere. And so, as Anjan said, you know, some of it comes from Nvidia, some of it comes from corporate venture. But these companies like Mercor, one of the ones in our portfolio, uh valuations went founding 30 million, 300 million, two billion, 10 billion. And how what what time? Two years? Two years. So, first of all, the 10 billion-dollar number is unprecedented uh in eight or 10 years and you what used to be incredibly rare is now
[00:10:00] incredibly abundant but the amount of capital going into the sector way outstrips the venture funds. And so what we generally see is the corporate money, the Nvidia money comes in to fill the void. But the people working there on stripes, they they say, “Well, this is really fun. I’m glad I made that Anthropic investment, but I’m going to go do my own fund.” So they the talent tends to eventually come out of the corporations and go into the two and 20 private sector to fill the space. So I think that that trend is now drawing in a huge amount of money, which is why we’re talking about it on this stage in Saudi Arabia. That the untapped but mobile capital is here in this room and if it jumps on the opportunity, the it’s like an opportunity I’ve never seen before. >> We can we talk about the two sides of AI? One is the buildout of AI infrastructure, right? And the other is AI applications. And the buildout of those applications, where do you see the capital split between those and the attractiveness to
[00:11:02] venture funds or public markets for those two things? Onj? Uh that’s it’s a it’s a really interesting question because the last few years, basically three four years were dominated by the infrastructure buildout, right? So most of the capital that was going into startups was being converted directly to GPUs. What’s interesting now is you have a whole category of super exciting application businesses. You know, where’s I’m I’m we got I’m just right here building one, right? In the coding space. And to build application businesses like that, you not sometimes you need GPUs, but other times you need tokens from other foundation models. That’s That’s now a raw ingredient as well. So the capital stack was just raw cash, then came you’d you’d convert raw cash to GPUs and then the the foundation model teams converted the GPUs to tokens and that’s an input now into application developers. Which is if you think about it, they’re more of a scarce resource. High quality tokens from foundation models is a much more scarce resource
[00:12:01] than raw GPUs and GPUs are a much more scarce commodity than raw cash. And so that’s the pref stack I would say of compute. >> you see the demand for infrastructure buildout continuing and and accelerating or topping out? Accelerating and not being able to not accelerating fast enough because now the fundamental constraint is energy. Right? We literally just don’t have enough power density in most of the legacy data centers in in most regions of the world. And you’ve got to go retool these data centers for GPUs. If you look at the new Blackwell from from Nvidia, you know, all all the research scientists I talked to are really excited because it’s got it’s got the NVL 72 networking stack, which means you can do a bunch of great big memory intensive training runs like video models. And then you get down to the brass tacks of when can that data center actually go live? When can we get it cabled? When can we get the energy permits? And that’s way after when the chips can actually get there. And so the
[00:13:00] infra span the infra needs are largely driven by demand forecasting. As we discussed earlier, demand is completely uncapped. And meanwhile, the compute supply chain is caught up with the energy the constraint hasn’t. The energy supply hasn’t. And so what we’re living through right now is this frenzy for energy contracts where compute providers are trying to outbid each other to buy literally just energy [clears throat] supply. Yeah. So the depending on which part of the infra stack you’re talking about, I don’t see things slowing down from a from a from a funding perspective. Like the the CAPEX going into this into infra is not slowing down, but what we may be faced with a hard wall on is just energy scaling. We just don’t have enough electricity to power the chips. Bonnie, what are you what are you seeing in the public markets in terms of energy, data center buildout, chip buildout, application companies? Well, it is all of the above, right? Um But I I do want to make a slightly more nuanced point. I I think at the moment
[00:14:02] the money that has been put into AI, you know, the two billion a day, a lot of it is probably put into the you know, these different opportunities on the premise that there is a promise that somehow it’s going to translate into things which are much easier to evaluate, right? So at the moment people just want these of AI. They don’t care whether it’s infrastructure, applications, energy, does not matter. But eventually, I think as the journey continues, I see a point where people will start to be a little more focused in terms of how we put a value on all these different opportunities. So from my vantage point, for example, and I think you raised a very interesting point. The energy bit is the is a million dollar billion dollar most a billion dollar question cuz without that, you really cannot go that far. >> Right. And therefore, if I look at my pipeline, for example, I think China as a lot of you know,
[00:15:02] has been quite advanced in terms of coming out with new energy solution. And it’s not only generating that new energy, it’s storing. And you know, I mean China’s a massive country, right? So how do you make sure that you know, you have all the grids talking to one another and then you can generate, you know, with the western and the northwestern part of China abundance of sunshine, wind and everything, right? You have the geographic or geo logical conditions to to help generate that that green energy. How do you make sure that you can disperse that, right? And to data centers again at every corner of the country so that you know, you can support all the data center, the infrastructure and all that. Now, with that as the building block, you therefore can proceed to the next level and talk about the compute the the applications and all that. Again, I would say that China has an advantage because it is still you know, a very big and dominant manufacturing hub. And and with that, it’s actually quite easy to think about possible
[00:16:01] applications and you know, you know, how you embed AI into production processes. I would also say that where I’m seeing a lot of activity is really the the data intensive sectors, right? So just to cite an example, we are now beginning to see um a lot of companies, you know, in the drug discovery business, for example, right? Embedding AI, which is as you could imagine, right? The the traditional way of drug discovery, you have to go through clinical trials, you have to select samples and you know, and and all that is data intensive, but if you can speed it up, right? With AI, you can imagine you are going to accelerate the pace of drug discovery so much. >> You have a friend of mine going public on your exchange in Silicon Medicine in the next I’m not allowed to comment on any specific um yeah, well, anyways, but I think you see my point there, right?
[00:17:01] Any data intensive business will be a darling, you know, in this regard. David, you’re seeing companies at inception. You’re seeing entrepreneurs, brilliant entrepreneurs. I think you’ve commented that the number of startups coming out of MIT and Harvard in the AI world is like quadrupled in the last few years. Yeah, more than What what kind of distribution what are you seeing? Where are they going into? Application layers, compute? What are you seeing as the the categories? The the companies coming out of MIT and Harvard are overwhelmingly going in into vertical use cases and then also some foundation model companies like Liquid AI will be on stage right after this. So there are a few of those, but many many more vertical use case companies. And the success rate of those is near 100%. And so they’re attracted to first they’re not super capital intensive. >> 100%. Well, so far for us, MIT and Harvard teams that fit a profile are 100%. I’ve never seen anything like it
[00:18:00] before and it’s because the use cases are so abundant relative to the talent pool. So if you have the talent and you’d have to be crazy to go after a bad use case right now. There’s just you can use AI for so many things. It’s very very different from crypto, which was the last wave, more similar to the internet. Yeah. The internet is incredibly flexible. You can use it for many many things and you saw, you know, when I started investing in the late 1990s, everything you invested in succeeded. Why? Well, because the internet can do almost anything. And so unless you’re insane and going after something really dumb, you’re going to succeed. So I haven’t seen that again in my lifetime until now and now it’s the same thing and the the value is enormous and the teams are thriving every single time, but they’re really attracted to the vertical use cases cuz they’re not as capital intensive as building out an entire data center. Now there there you know, Chase Lochmiller is doing Stargate. So there’s one guy who’s an exception to that. There’s a $500 billion buildout. So but that’s relatively rare. Most people go after the use case. >> And how quickly are you seeing the
[00:19:00] valuations in those kind of company scale? I mean it’s it’s in like Chase Lochmiller or like >> Well, no, in the in the companies in that are doing the vertical Uh in the Link Studios, yeah. >> typical entry valuations are what they’ve always been, maybe 20 30 million dollars. First funding will be 100 to 300 million and then within two years, if you’re going to be a unicorn, you’re going to get there in two years now. Which means the founders now are still 23 24 years old. So that’s a new thing in the world, too. You know, we have a bunch of people that I can name. You know, I think about my entire lifetime of investing, can name like three or four people that I knew or invested in that hit billionaire under the age of 30. Now I can name eight that we’ve invested in in just the last few years. Yeah. So it’s like so there’s this new class of person roaming around that barely has a driver’s license, but has a billion dollars in liquidity. And so So have to kind of adapt to that. >> billionaire, being a billionaire was a big deal. Now we’re just going to wait for the trillionaires to start. Well,
[00:20:00] we’re all born in the wrong age. Yeah. Yeah. Yeah. You know, I I want to understand what you guys consider the biggest risks uh over the next year. Is it compute cost inflation? Is it talent scarcity? Is it regulatory intervention? We’ve been on this incredible inflationary and exponentially growing curve on all things AI. Uh just like you used to be add.com on the end of your company, now it’s like, oh, we use AI. Uh Anj what are you seeing as the risks? Um so on fundamental like progress of capabilities, we already talked about the one energy, which I’m concerned about. But So double double double click on that. So will will these companies have access to sufficient electrons to run the data centers? Is it is it what what is the scarce resource in the in the chain? Um in the United States, I think that’s a direct function of whether the
[00:21:00] permitting regulations that the current administration is working on end up getting executed on. So there was a big plan that was introduced, the AI action plan about 2 months ago, which I think was a fantastic start. And if you if you go sort of line by line through that, it really is a very precise, methodically laid out document that says, here’s what we need to do to unblock progress. And I think if we can operationalize it and execute it, then we should be good. But rarely has that ever happened at scale without a ton of um bureaucracy. A ton of bureaucracy. And and this is my second actually concern, which is without a ton of I think civil blowback. Because the reality is putting these massive data centers down, cabling, reallocating parts of our power grid from other things results in tough trade-offs we’ve got to make as a society. And and I just I want to respond to the previous um point a little bit where it it it is true, we are seeing enormous wealth creation amongst this generation. Right? Anthropic has gone from a company that was you know, a couple hundred million in valuation just 4 years ago to 183 billion dollars in 48 months. But I
[00:22:01] don’t think we should be celebrating that as much as we kind of are right now because at the end of the day, the public is not participating in that wealth creation. The vast majority of wealth being created by frontier AI is locked up inside of private capital like our funds. It’s it’s locked up inside a small group of talent that is super mission-oriented, but I don’t think we’ve really figured out what happens when the rest of the public goes, well, where’s my piece of the future? Yes. And I don’t think we’re ready. I don’t think we’re talking about it enough. And I don’t think governments are doing enough to realize how dire it’s about to get when 30% of your IT services GDP sector gets vaporized by tokens. If you’re India, for example, where double-digit percentages of your GDP are literally IT services, what do you do when Claude and GPT-5 tokenize like vast portions of that flow? We I think we love to talk about productivity growth, and we don’t talk about how to manage
[00:23:00] the short-term transition pains. And that’s going to be ugly. >> So that you’re adding that to our risk profile, civil unrest. Absolutely. Well, good example of that, too, is you know, just a few months ago when Sam Altman said, “Hey, I’m going to give everybody in the company a $1 million retention bonus, everybody.” And the intention was for that to be cool. The reaction worldwide was, that’s not cool. And so now you’re seeing the AI leaders, it comes up on the Moonshots podcast a lot, the AI leaders are really downplaying the rate of progress uh because the people that are picketing outside the door at OpenAI headquarters are lined up a deep now. Yeah. And they’re like, look, all this wealth that you guys are all billionaires, what about everybody else out here on the street? And they don’t they don’t need that. >> it’s, you know, just to put a finer point on that, I know a number of the technology leaders and investors in Silicon Valley who have been getting death threats. Yeah. And then they lock down their companies, they lock down their homes. Uh and this is before we
[00:24:00] we’re seeing CPI of electricity going up, but this is before we’re seeing the real uh layoffs that will occur. Well, so I I I I think this is important. I think AI is going to get blamed for a lot of layoffs that have nothing to do with AI. A lot of the layoffs we’re seeing today from big tech companies are really just people correcting over-hiring during the serf era of 2010 to 2020. >> Well, also the the print money era of of of COVID. >> So the easy money’s gone, and a number of big tech companies that just thought they could keep putting chasing returns by over-hiring, which was a fairly rational thing to do then. >> In fact, the government was paying you to go higher. >> Exactly. Right? And but but the incentives have changed. So one, I just want to there will be a lot of boogeymanning around AI that has nothing to do with AI. >> Agreed. Okay. But once we’re through that era, what happens is people are going to start asking, why aren’t my why isn’t my pension fund, my sovereign fund, my retirement plan participating
[00:25:00] in the AI wealth creation opportunity? Mhm. And that’s why I think to the point of this panel, which is how do we fund the future of AI, we should be asking, how do you connect the frontier AI growth to public wealth creation? Mhm. And there’s a bunch of institutions whose job it is to steward our wealth, sovereign funds, pension funds, state funds. Why aren’t they investing on the cap tables? Yeah. Why is it family offices? Why is it high net worth individuals? When we went out to raise the seed round for Anthropic, I made 22 introductions to them up and down Sand Hill Road. They got 21 no’s. So we had to scrape together 100 million bucks, which sounds like a lot of money, but which was a lot more money then back then. Now actually to to to to David’s point, it may not be that much, but really that funding round had to be pieced together from angels and high net worth individuals. And I’m still shocked at how often today traditional venture sovereign funds, traditional pension funds are not being aggressive enough in managing the steward taking their job as a steward of public capital and exposing it to frontier AI wealth creation. It’s just not happening fast enough. Dave, do you
[00:26:01] want to add on the on the risk side or what? >> Yeah, yeah, well, I completely agree with what Anj said, and I’ll give you another parallel risk, which is that, you know, the the core AI companies that do things like customer support, white-collar automation, just killing it. I mean, adding immense amounts of value. And the investment community coming in has started to extend that to, oh, tech is a good place. Let me put 200 million into fusion energy. And they’re like, well, that’s not AI. Well, but it’s going to create the electricity about 4 or 5 6 years from now to fund to So it’s related to AI. I’m like, well, okay. But that’s very capital-intensive, and you’re not sure it’s going to work. And so I think it will work, and I think it’s a good area to invest, but if it doesn’t work, that’s where you’re going to have this what happened to the internet in 2000. The internet was very real, and if you waited long enough, it came roaring back, but everybody lost confidence in 2001. Why? Because of some really bad peripheral investments. And now we’re seeing that, and I don’t want to throw too many things under the bus, but some
[00:27:00] things like uh robotics is very capital-intensive, fusion energy is very capital-intensive. It’s not the obvious win of AI. It’s a peripheral investment. Some of those will be good. Some of them are going to consume a ton of money and turn into losses, and that may scare off the entire investment community. And so that that would be tragic, because if you look at things like just if you look at AI voices doing sales and customer support, that’s half a trillion dollars of payroll worldwide today. The AI does it better than anyone on the phone already. Like it existing We just need to deploy that half trillion. If you invest in that, you cannot go wrong. But if you get sold an investment in something that’s kind of like, well, quantum computing also might work. Maybe. Maybe it will, maybe it won’t, but much more speculative and very very capital-intensive. Please. >> I do want to chime in there. I think um I’m listening to all this, right? I mean, one part of me is saying, I want to democratize these investment opportunities, right? Let more people
[00:28:00] partake in the in the party. But on the other hand, right, just given, you know, how the the current ecosystem has built out, the valuation, right, has has already, you know, you know, hovering somewhere up here, um opening the door for, you know, investors, especially retail guys, to partake in in the public markets also caused me concern, right? Because, you know, I mean, for all I know, they could be the last one, right, being handed the um um you know, being the last ones in at the party before the whole thing collapses. So I think, you know, I would call that as a risk, right? How do we actually find that new equilibrium where these opportunities are not just monopolized by a very small group, right? How do we make more sense out of the valuation which we are seeing, which is again, right, being established by a very small and rather opaque in some instances price discovery mechanism. Um but, you
[00:29:00] know, to your original question about risk, I do see the energy piece as one which is very difficult to solve. Because, I mean, even at my company, right, we’re exploring with, you know, what we can do with AI. Um we have come up with a few cases, right, that we we we were experimenting with. And the next thing you know, I come you know, the elec- the electricity bill arrives, and you started scratching your head, right? I thought AI is going to help me with productivity and make things faster, easier, you know, more accessible. Yes, but there’s always a cost there, right? So I guess, you know, people just need to >> have a minute left for closing thoughts from each of you. Anjaney? I think the answer lies in institutions who who represent the public, sovereign funds, wealth funds. You’re right, opening up the markets to retail investors who may not understand what’s going on may not be is not the answer but I think institutions who represent the public are the answer and it’s our job to educate them and make them more aggressively I think take a position in the wealth creation opportunity that’s happening otherwise the public will get
[00:30:00] left behind. Bonnie closing thoughts on who’s going to fund this? >> I agree I will agree with that. I I really think that you know everyone in this ecosystem need to work together to find that new equilibrium. It shouldn’t be wealth creation for a tiny fraction of the world’s population um and we need to find the right right way to get it done. All right Dave. I think the most important thing that I heard on this stage today was what Andre’s Andre Andre saying the story of how Anthropic got funded. So many people are not getting in the game and Silicon Valley investors that are just walking down the street and investing each other are killing it and running away with all the all the gains because it’s just not that hard. You just need to get into the loops get into the places that are making these investments and get in the game and then the pro rata rights on that deal alone would have allowed you to invest a follow on of probably what four five ten billion dollars of follow on. But you just had to be there in the game at the outset. Every week my team and I study the top 10 technology metatrends that will transform
[00:31:01] industries over the decade ahead. I cover trends ranging from humanoid robotics, AGI and quantum computing to transport, energy, longevity and more. There’s no fluff. Only the most important stuff that matters that impacts our lives, our companies and our careers. If you want me to share these metatrends with you, I write a newsletter twice a week sending it out as a short two-minute read via email. And if you want to discover the most important metatrends 10 years before anyone else, this report’s for you. Readers include founders and CEOs from the world’s most disruptive companies and entrepreneurs building the world’s most disruptive tech. It’s not for you if you don’t want to be informed about what’s coming, why it matters and how you can benefit from it. To subscribe for free, go to demandis.com/metatrends to gain access to the trends 10 years before anyone else. All right, now back to this episode. >> [music]