Episode Transcript
[00:00:00] Speaker A: The asset owners need to innovate also a little bit, adding at least one person, one talent on their team that is more prepared to understand this type of solutions. You know, it's for AI and it will be more so for Quantum because otherwise you risk staying on the sidelines and, you know, just don't act upon the innovation that I'm guessing will come here very soon.
[00:00:32] Speaker B: Welcome to the Institutional Edge, a weekly podcast in partnership with Pensions Investments. I'm your host, Angelo Calvello. In each 30 minute episode I interview asset owners, the investment professionals deploying capital, who share insights on carefully curated topics. Occasionally we feature brilliant minds from outside of our industry, driving the conversation forward. No fluff, no no vendor pitches, no disguise marketing. Our goal is to challenge conventional thinking, elevate the conversation, and help you make smarter investment decisions, but always with a little edginess along the way.
Today we're talking quantum technology. And yep, I can already hear some of you thinking, isn't it a little early for this? Well, maybe. But here's what changed my mind. After I briefly touched on the topic with Bettina Kitsler in an earlier episode, the listener response was clear.
This is something institutional investors know they should understand, but haven't had a good jargon free introduction to. And my timing of this episode isn't arbitrary. Quantum computing attracted nearly 4 billion in VC funding in 2025 alone, and projections for this year are higher. So a growing number of asset owners are gaining exposure to quantum simply through their VC allocations. Which means, whether you know it or not, quantum quantum tech may already be in your portfolio. And also banks and asset managers are among the earliest institutions running quantum pilots. So today we're going to try to demystify it. What is quantum computing really and what can't it do yet, and what should institutional investors actually be paying attention to? Help me work through those questions. I've got two people who know this space deeply, Dr. Daniel Volz and Dr. Elisabetta Basilico. Plenty of PhDs on the show today, which means I'm counting on both of them to keep it accessible. Dr. Volz is a deep tech entrepreneur and quantum expert. His experience spans strategy, consulting, industrial R and D and venture building. He's the founder and former CEO of Quipu Quantum, a quantum software company focused on developing application and hardware specific quantum algorithms for real world industrial problems. Prior to founding Quipu, Daniel worked on quantum computing strategy and applications at McKinsey and Company, where he advised global clients across chemical, pharmaceutical, energy and finance on the commercial potential and the realistic timelines of quantum technologies. He later joined BASF and he was contributing to enterprise level quantum initiatives. Daniel's background combines hands on scientific research with business execution. He holds a PhD in chemistry from Karlsruhe Institute of Technology where his research focused on advanced computational and physical chemistry topics. This foundation enables him to translate complex scientific concepts into practical technology strategies and scalable products, which is important for us in this space. His perspective today is that of an operator.
Where is the value really accruing? Who are the key players and where are the investments going?
Dr. Elisabetta Basilico is a financial professional with over 20 years of experience working with institutional investors, including pension funds, asset managers and family offices. Her areas of expertise include asset allocation, risk management, fund due diligence and performance analysis. She's the author of Smarter How Academic Insights Propel the Savvy Investor and she's written numerous articles on investing and financial education. She earned a PhD from the University of St. Gallens and has held a CFA since 2007.
We're going to put their full BIOS in the show notes now. Let's get on with it. So Daniel and Elizabetta, welcome to the show.
[00:04:31] Speaker C: Thank you Angelo, it's a pleasure to be with you today.
[00:04:34] Speaker B: Elizabetta, is it a pleasure for you too?
[00:04:36] Speaker A: Yes.
[00:04:39] Speaker B: All right, well let's get to it. And Daniel and Elizabetta, as we discussed in our pre call, you know, what we're trying to do here today is demystify quantum tech. So our audience, institutional investors, they get a better understanding of this because it's front and center for them now, whether it's in their VC funds or if they're seeing their asset managers even starting to dabble in at some of the hedge funds. So Daniel, I want to start with you. Pretty simple. What is quantum computing really?
[00:05:07] Speaker C: Quantum computing is essentially a new computing paradigm that uses quantum physics at the heart of the computations that are being done, which is different from the conventional physics rules that are dominant when we are calculating using bits essentially in CPUs and GPUs.
[00:05:25] Speaker B: That's it.
There's got to be more to that sentence than that.
[00:05:32] Speaker C: I think you need to let this sink in because what I said is that it's really a completely different compute paradigm. It operates on different physical rules. And of course that implies that you can do different things using quantum computing. So oftentimes people simplify and say, you know, this is just another faster computer, like you know, next GPU generation, next high performance compute generation. But I think there's more to that because using the Rules of quantum physics, especially, especially phenomena like entanglement and superposition, really allows these machines to do certain operations, not all of them, but certain operations in a significantly improved version versus what we can do with CPUs and GPUs. So that really opens the doors to doing things quite differently. And of course I'm not saying that, you know, there will just be a switch and at some point everybody will have, you know, a quantum computer in our mobile devices. It's going to be more, more nuanced than that and it's essentially going to broaden the scope of processing units that we are using in the industry and for other applications.
[00:06:40] Speaker B: So it's built on a different paradigm to start with. But I mean, given your comments about cpu, gpu, does it use different hardware?
[00:06:48] Speaker C: Yes, exactly. So sometimes, as I would say in the industry, in the niche of quantum computing, but I think this term becomes more abundant. So people say of, of course CPU and GPU and sometimes people say, okay. There are also QPUs, quantum processing units.
That is really different from a hardware perspective. I have to say that we are still at this BHS vs. Betamax phase in quantum computing because there's not one way of building QPUs right now. There are a lot of different, fundamentally different hardware modalities out there like VHS and Betamax, which have been different like 30 years ago. And different companies are driving different core technologies.
And we are not working with bits anymore. Essentially we are working with so called quantum bits or qubits. And there are many different ways of building qubits right now. And you know, I don't want to be too technical, but it's, they are really orthogonal approaches right now about four or five that are driven by a very interesting ecosystem of hardware vendors that are driving these different efforts forward.
[00:07:54] Speaker B: Will there be some type of winner in that? Because we know that Betamax didn't exactly kick ass. It was VHS that won that one. So.
And also this must be something what we're seeing VCs play in as well, you know, because there's gotta be funding for this and some of it may come from the government, but I'm assuming some of it comes from the private sector.
[00:08:14] Speaker C: Yeah, definitely. I would say the private sector really picked off its investment in the last year or two. And I think there was a, there was a VHS Betamax war for some time in the past.
So it was not an immediate one punch knockout for vhs. I do think that there will be an ongoing struggle for the various QPU modalities probably in the next five to 10 years. So I don't see this becoming a winner takes all kind of game in the next 10 years or so. So there will be cases where one type of hardware will be significantly better and therefore the better economic choice than another type of hardware, depending on the application.
[00:08:56] Speaker B: Okay, so the hardware you're saying is going to sort itself out over time.
[00:09:00] Speaker C: It's going to sort itself out, but I would say the next years that we will see a coexistence of different, of different modalities. So I don't think that there will be a convergence in, in the next, in the next three, four years or so.
[00:09:12] Speaker B: Okay, so we've got different hardware and you're talking about a different paradigm because we're talking about, I want to say qubits rather than bits.
[00:09:22] Speaker C: Exactly.
[00:09:23] Speaker B: Bits to me are 0 and 1, right? Qubits. I mean, it's got to be different than 0 and 1 in my simple mind. Am I right?
[00:09:31] Speaker C: So qubits can be 0 and 1, but they can also essentially assume so called states that are between 0 and 1. And I think this makes it a little bit complicated because in the purely classical digital world we're thinking, okay, ones and zeros easy, or, you know, not that complicated. Not super easy, but it's not that complicated. I think in the quantum world we have to get used to having a lot of, into a lot of concepts at play that are not that intuitive.
So quantum physics is fundamentally different. I think even people like, you know, Albert Einstein have been, have been struggling to learn to kind of understanding and accepting some, some of its rules which are not that intuitive. If you're just a human being deriving how nature works from your day to day experience. And I think the fact that qubits can have many states between 0 and 1 is one of these effects and that ultimately allows these computers to do things that are not possible for the CPUs and the GPUs who are either 0 and 1. And then this is on a technological level, the source of potential outperformance that quantum computers can bring.
[00:10:40] Speaker B: Well, I mean, if Einstein had a hard time, I don't feel too bad. I don't know about you, Elizabeth, but you know, let me go back even farther, Daniel. Where did this come from? I've been in the AI space for years. I understand the history of traditional AI, but you know, I know where it came from. Where did quantum computing come from?
[00:11:00] Speaker C: It has been proposed on, as a theoretical concept, so some decades ago by essentially physics academics. And I think conceptually it was born from frustration because people realized that certain computations, especially if you want to calculate, predict nature. Chemical systems, physical systems, some of these equations are incredibly complicated and very difficult to solve. And that stems from the fact that nature, especially on the microscopic level, uses the same effects that quantum. That quantum computers are using. So you, the same way that quantum computers can access computations that are not accessible for CPUs and GPUs can also be turned around. So if you want to simulate certain things, you will end up with equations which will be extremely difficult if you're just using bits that can be between 0 and 1. So I think that that is kind of why people have been dreaming for a very long time, since the 80s, essentially early 80s, about using quantum physics to make computations. And then over the last decades, it has been slowly turned from, let's say, academic fever dream into real technology, which is now slowly becoming commercially available.
[00:12:17] Speaker B: Was there a catalyst for that transition from theoretical to practical?
[00:12:22] Speaker C: I would say one of these catalysts has certainly been IBM as a company, because some 10 years ago, they essentially released a quantum computer with a very low number of qubits. So not, not something that you can use to, you know, fix climate change or cure cancer or do sophisticated financial calculations, but they made a quantum computer accessible via the cloud. And then I think they even gave away certain amount of minutes per month for free for everybody that was signing up. So suddenly there was this new compute paradigm readily available for everybody that wanted to go through the gauntlet of learning how to use it. And I think this event some 10 years ago was really probably the most relevant catalyst in recent days. And of course, this is, I think, an economic catalyst. All of that would not have been possible without a lot of scientific breakthroughs. I think it's hard for me to say that there's one or two individual scientific breakthroughs. This has been a series of scientific breakthroughs from many different teams in the last 10, 20 years, which now have led to quantum computing being kind of at the threshold of industrial usefulness.
[00:13:36] Speaker B: You're talking about how it was. It was really a research project that's now, I'll say in the lab.
What are the research hotspots for this? Universities, I assume, are playing a big role in this. Is that a fair assumption? And if so, which universities?
[00:13:54] Speaker C: You can point out to a lot of universities, I would say that. And of course, universities are still relevant to solving some of the problems and kinks that quantum computing has. But I would say that that that really, with these IBM releases in the last decades, this for me marked the turning point where universities still became a relevant player, but it really turned into something where industrial players are now slowly moving to the driver's seat. And of course, if you want to name universities, I think that has been proven by global community. A lot of leading universities in the United States, Canada, Europe, also Australia, Singapore, China, Japan have been driving that. But I think there's not one hops hotspot that is driving all of it. So it's essentially a collective effort where a lot of universities are driving different pieces in the stack. And now the commercial parties, like the Googles, like the IBM's of the world, like the startups that are out there, are slowly taking these pieces, refining them and turning them into commercial solutions.
[00:14:55] Speaker B: Daniel, you said that with quantum you could do things that are not possible with what I'll call traditional computing. With bits. What are some of the things that can be done?
[00:15:07] Speaker C: A mature quantum computer can essentially do tasks like sorting and finding, finding solutions in a vast solution space in a way that is not possible for conventional computing. So if you want to find, if you, for example, want to do route optimization, of course there are heuristic algorithms which, you know, if you're interested in like 90% of the best solution, then you know that is fine. But if you really want to go, you want to find the best solution, the single most optimal solution, then this is something where a conventional computer would need to go through every single potential route, for example, to kind of find the shortest or the best one. And that number can be quite large if you have a few nodes. So there are some kind of, some problems which really scale horribly exponentially, some even worse than that with a problem size. And then very quickly you end up with problems which seem like solvable and in reality not really solvable, if you don't want to use heuristic algorithms. So that can be used for finding quicker routes, finding optimal resource allocations, but also for things like, as I said, predicting nature was one of the motivations for going into quantum. So there are cases where you can predict the properties of chemical reactions of materials in batteries, for example, chemical process catalysts, predicting how a certain molecule that maybe can be used as a pharmaceutical drug to treat certain illnesses would need to look like on a molecular level. So these are some of the applications which are done altruistically, semi empirically right now, but still are not yet as we would like them to be.
[00:16:55] Speaker B: You mentioned there are some limitations and I'm not just talking about, you know, the ongoing research but are there are there inherent limitations to Quantum? There are just things that another it can't crack.
[00:17:08] Speaker C: I mean there are some things which it could crack but would maybe not be meaningful to crack. So if you just want to add numbers, for example, you know, even if the numbers are very large, you know, a billion plus a billion plus a billion, something like this is something which quantum computers could do, but are not inherently better than CPU or GPU better based solutions. So there will be cases where it's just economically not meaningful to just use a quantum computer. And I think this is comparable to, you know, fundamentally a jet plane could take you anywhere, but if you just want to go to the grocery store, then maybe you walk or you take a car and not jet plane. So I think this is, I think a good way to think about some of the inherent limitations where it's just not meaningful to use a quantum computer. Of course, quantum is not yet a mature technology. I would say quite the contrary. It's right now at a point where there are first applications where, where quantum performs incrementally better than CPU or GPU based solutions. So it's, it's at a point where it starts to become more powerful than, than other benchmark technologies. But of course it, it still can improve and will improve by, by quite a lot.
[00:18:20] Speaker B: Okay, so I understand the use cases at a high level in what you're telling me is it could solve very complex problems better than the current alternatives, some of those related to finding the optimal solution instead of just a 90% solution. Yes, I mean, what can't it do yet? I mean, because you've indicated that there's going to take some time for quantum to reach what I would say, maturity. You used the word it's still immature. Until it matures, what can't it do?
[00:18:51] Speaker C: So one of the things which, which oftentimes gets mentioned when people talk about quantum computing is breaking cryptography. So conceptually quantum computing is capable of breaking things like RSA encryption, which is, you know, one of the foundations of, of the encryption of the Internet of cryptocurrencies and so on. And fundamentally the RSA encryption relies on the fact that it's that very difficult for more conventional computes to do prime factorization.
So if you want to figure out what are the primes that make up numbers like 12, you know, okay, 12 is 3 times 4 and 4 is 2 times 2. So it's 3 times 2 stuff that people let children learn in school. But once you reach a certain length for the numbers, it becomes not so easy anymore. And once you have a certain length of a number, you can essentially be assured that no conventional computer can figure out the prime factors. And that provides a level of security that we are using in encryption. Now, once we, once we would be able to do 2048 bit integers with quantum computers, this would pose a serious security threat because then suddenly the current encryption standards would no longer be secure. And that kind of calculation is fundamentally possible with quantum computers. There have been scientific papers about the date back 30 years or so, but the current quantum computers are not yet powerful enough to do, for example, 2048 bit integer factorization.
So that is one example, for instance. And of course I mentioned that Quantum is now at the point where it starts to solve first industry use case. You will, as an industry user always be able to say, okay, if I have, if I give you a problem that is 10 times larger, will you then still be able to solve it? And in many cases still the answer will be no, you will need to wait until next year's hardware generation perhaps. So it's really kind of that kind of battlefield right now.
[00:20:49] Speaker B: I have like one final question before we turn to the investment side, Elizabeth, and that is, you know, I've read some of the papers and it talks about how Quantum could actually improve what I would consider to be classical artificial intelligence or could be used with machine learning.
What's the nexus there between quantum and. I'll just use the general term machine learning.
[00:21:12] Speaker C: So I first want to talk about one notion. Oftentimes people think that there's an inherent competition between AI and Quantum. You know, people say, isn't AI going to eat Quantum's lunch? Or isn't Quantum going to eat AI's lunch? And I would say it's really going to converge in many areas.
So I do see the future being comprised of workflows where you would have a machine learning workflow, perhaps using CPU and GPU based steps, but then maybe also having a QPU based step in it. And I think some of the examples are for instance, getting more insights from limited data sets. So you know that AI is very good. If you have a large amount of very high quality data. This is the typical sweet spot where AI gives you excellent answers. But if you want to apply it in an area where data is scarce, or you know, you don't have high quality data, or it's super expensive, then I think this could be. There are great opportunities for Quantum to actually be applied in these areas. And some of last year's claims of Quantum outperforming CPU and GPU based solutions are exactly on that nexus where AI workflows do exist, but you know, they are not good, not yet good enough, or, you know, you would like them to be improved. And I think there are a lot of applications like making better results with limited data sets, image classification, for example. Another example is another example. Or creating machine learning workflows that are using certain Quantum operations. So I think there's a lot of great, interesting stuff to do there.
[00:22:45] Speaker B: And is there a lot of research in that space already?
[00:22:48] Speaker C: There is a lot of research in that space, yeah, I think there's. Yeah, that's a fair answer. I think that there still needs to be more. So more work still needs to be done and can be done. So I would not discourage any young researcher to go down that route because it's kind of already fully explored. But there's also a lot of work that was done in the past.
[00:23:07] Speaker B: And I'll kind of conclude this section by asking, give us an example or two of where Quantum is actually being used today in some type of a commercial setting.
[00:23:18] Speaker C: So right now these applications where Quantum is being used in production environments are super few yet. So there are not many of these applications. Applications. I would say where it will be first used is probably in machine learning workflows and combinatorial optimization problems where you are looking at problems where small gains in optimality, slightly better answers are very valuable. And right now Quantum is already a billion dollar opportunity in the sense of that is actually the amount of money that large corporates are right now spending on Quantum. So it's more than 1.5 billion as of last year, but it's not yet in production. So we are kind of now at this phase where Quantum has been proven to outperform solutions and first companies are seriously putting it into productive environments. But it's not yet at a point where you could say Quantum is generating large measurable profits or gains. So that is something that I would expect in the next couple of years.
[00:24:18] Speaker B: 30 episodes in the Proof of Concept has been established.
Guests like Don Pierce, Venus Phillips, Andy Green, John Grable, Paul Greff, Brad Brindle, Mark Steed, Leo Swavoda, Joe Eppers, Rene Diresta and Greg Brown. Hey, they don't come on a podcast that hasn't earned their time.
So we're ready to take the next step. We're now accepting sponsors for the institutional edge. And before you ask, here's how it works. Sponsors support the show show, but don't get editorial control over the content and no, you will not be hearing me reading ads on energy drinks and tax preparation services. This is still the same show. If sponsorship interests you, contact Julie Parton at Pensions and Investments. She can tell you about all the benefits. Her contact details are in the show notes.
Elizabeth Bennett, you're here because of your deep background in investing and we've got your bio and I'll tell people all about you in the introduction. But I wanted you to represent the audience today and have you ask Daniel questions that you think a pension fund, and you worked at a pension fund for years managing a portfolio even in the US I was surprised by that when I met you years ago.
You understand family offices, you understand endowments, you understand, you know, the problems that these investors are trying to solve more and more with technology.
What questions would you be asking Daniel, given what we the little I know about Quantum? What would you be asking him, please?
[00:25:55] Speaker A: Thank you, Angelo. Yeah, so before I ask my first question, I want to say just a small premise going back to the qubit definition because Daniel and I wrote an article and in that article we tried to explain that quantum computing with the qubits can actually solve problems much faster. And we use. So I'm going to see if maybe the audience understands this a little better because we use an analogy. And so we use this analogy of searching a phone book with a million names. A classical computer would check names one by one, whereas a quantum computer can potentially explore all the names at once. So that's where the speed comes in. And linking to this, I want to ask this question to Daniel because a lot of portfolio construction, which is a core step in building a portfolio, already relies on pretty sophisticated optimization techniques.
Apart from the speed, where do you see Quantum fit in Italy? Will it at some point be able to help us improve how portfolio are built? Or is the opportunity and the value somewhere else from Quantum?
[00:27:15] Speaker C: I think this is a good, this is a great question.
It depends a little bit on which kind of portfolios we are talking about because there are some portfolios where you would, where time is of the essence and you know, building a portfolio that is, let's say 80% or 90% optimal in a split second is, is the value case and that's the bar. So if you take longer but get a better portfolio, you will not say you will not find it useful. But of course there are certain applications where you have where, where you are a little bit more generous with your time and where you would really appreciate if you go even just from 97 to 98 or 99% optimality because multiplying that gain with a large portfolio equals a large economic outcome. So I think Quantum can fundamentally be applied on both facets of this problem. It can be used to accelerate certain portfolio optimizations, but it can also be used to getting better optimality. What I've also seen in the financial industry is that people are introducing constraints, having certain volatility requirements.
I've even seen institutions look at sustainability requirements, you know, adding boundary conditions and boundary conditions. You know, I want the highest returns at a certain footprint. And of course that that makes it fundamentally more difficult for any CPU or GPU to solve these kind of things. So you could also use it to get, not to not get better or faster, but be more sophisticated, having more boundary conditions, having more restrictions in place to get to the same performance level under much, much higher standards, for example. So from my perception, these are the three step. These are the three things, the three fundamental areas where Quantum could be applied to a portfolio optimization.
And you know, what is the appropriate value really depends on what the end users are looking for, what kind of portfolios people are optimizing, for example, okay, so this is.
[00:29:12] Speaker B: Quantum offers the possibility of better optimization in terms of allocation of resources, whether it's financial resources or others. And you talked about different time periods.
Maybe it's shorter, maybe it's longer.
Elizabett, would you see Quantum as it comes, as it matures, to use that term again, to replace our current. Let's focus on SAA for a minute. The longer the strategic view, do you think it would replace or would it complement the asset allocation decision at a pension fund like when you were with Kathy Letito? What do you think?
[00:29:47] Speaker A: Well, it's a great question and you know, I'm going to answer with the help of Daniel hopefully, because what I want to say is I'm not going
[00:29:55] Speaker B: to help you know that. So go ahead.
[00:30:00] Speaker A: So, you know, one of my core jobs is actually building strategic asset allocation for institutional and family offices portfolios. And the challenge, there are different challenges when you're optimizing for a portfolio. So I'm very interested in how Daniel is going to comment on this.
Ideally and potentially I would love if we get to a stage where it will replace the current algorithms because it's true that they are sophisticated, but there are three main challenges that I see currently.
One is the forecasting of the inputs, returns, volatilities, correlations. So it's an art and a science. It's not just a science that we have currently from research. So that's one challenge. And then in performing the optimization, the challenge is that oftentimes we are approximating the risk preferences of investors. So for example, when you use the traditional mean variance framework in its basic or advanced forms, you're using what we call a quadratic utility function. That function majorly approximates the risk preferences of an investor because it says that they're, you know, whether you achieve a return less or more than a return, your pref, you know, your preference preferences, risk preferences don't change, they're symmetrical. And it's not true because an investor is more worried about achieving a target than getting an above average return. And this comes from behavioral finance.
So current algorithms don't solve for this. And then there is the whole constraints problem that Daniel already touched on. Because in reality when you perform an optimization is not unconstrained.
Different investor have different constraints. And so the more constraints you add, the more you're going to get a suboptimal portfolio. So this, there are three challenges that currently with the, you know, with the research that we have, with the tools that we have, nothing solves them that you know, all of them. So I'm going to then turn to Daniel and say, Daniel, do you see one area being faster for quantum computing, computing to solve? Will it solve all of three of the, you know, the. All, all three of them eventually. So I'm very interested about this too.
[00:32:35] Speaker C: I think ultimately it's going to be relevant for all of these cases. I would say the speed advantage, especially for very fast portfolio optimization will be a tougher nut to crack. And I say this because the benchmark to beat is really relying on heavily integrated solutions and of course integrating quantum into replacing some CPU and GPU based solution in a workflow that already exists and then be much faster is challenging because you have to invest a lot in the kind of integration work. So I would say that's possible, but maybe not the first application where it's going to be relevant. So I would say getting to better answers or having more constraints are probably going to be, are going to be relevant earlier than just sheer speed advantages, especially in the portfolio game.
[00:33:27] Speaker B: I wrote a paper published a few weeks ago about how you could use reinforcement learning, specifically deep reinforcement learning, to solve the saa, the Strategic Asset Allocation Optimization problem.
I'm not sure if it addresses the three points elizabetta made, but it seems to me it's a quantum leap. How do you like that guys? It's a quantum leap forward from just trying to use human intelligence to now using reinforcement learning. Which does a pretty darn good job in terms of solving nonlinear optimization.
Any comment from the two of you? Because, you know, I. I could always take that paper down. It's a preprint, basically, so I didn't
[00:34:06] Speaker C: read the paper, so I will not recommend to take it down without reading it.
[00:34:15] Speaker B: Fair enough.
Okay.
[00:34:16] Speaker A: Caught us unprepared. We didn't read the paper.
[00:34:19] Speaker B: That's all right. I do have one other question I wanted to ask, and I should ask this earlier. What kind of data are you using as inputs for these quantum algorithms? What's the data look like? Because I know what Elizabetta and I have grown up on. We use certain data to make decisions. It's pretty typical. But what do you do in quantum?
[00:34:36] Speaker C: The quantum is going to use the same types of data, essentially.
[00:34:39] Speaker B: Okay.
[00:34:40] Speaker C: I think quantum offers the chance to use data beyond that, especially if you're talking about applying quantum machine learning on quantum systems. But I think this is something which will be relevant in the future, where you would use quantum data as input for quantum machine learning. But it's more or less the same types of data because, I mean, if you think about it, ultimately the same people, the same domain experts and end users are going to use it. So I would say there's very little incentive for them to use Quantum. If the entry ticket is here. You need to learn another data type or transfer all of your stuff into other formats.
[00:35:17] Speaker B: Okay, Elizabeth, back to you. Thanks for giving me a chance at my own podcast to ask a question.
[00:35:24] Speaker A: I guess I have one final more general question for Daniel and for an institutional investor that wants to be cautious but not premature, what do you think is the best way to engage with this? Still feel today?
[00:35:41] Speaker C: I mean, there are a lot of companies going public these days. If you look, if you just count it, the majority of the public companies, pure play companies, are dominated by hardware. There are some specs right now happening that are also quantum software. Very few, I think. So if an investor wants to be cautious, which implies diversification, I would probably look at building a portfolio of different hardware modalities.
Coming back to the point I said earlier, it's right now not apparent yet that in the next decade there will be one clear winner. So I would say, just saying I have one big position in one of the quantum hardware companies is a tad risky because it could be that the investor selects a company that will maybe be obsolete in eight years or so. So building a portfolio is, I think, a good idea. And of course, there are for institutional investors also opportunities to invest into other funds. So there are a couple of quantum focused funds around Europe and the United States predominantly where people could essentially invest to say hey, I'm working with a team that maybe does the portfolio building for me, a venture capital fund for example, and maybe is smarter than I am as a non expert in differentiating between solid and not so solid cases in Quantum, I think the time to do so is right now and I think that is a good opportunity. Of course really cautious investors could also just look at the existing tech companies because companies like Google, Microsoft and IBM, intel also to some extent are also driving Quantum efforts. But of course none of them is really driving a broad portfolio. So the risk catching would come from the fact that Quantum would only be a small component in the overall business of activities of, you know, Google or Microsoft.
[00:37:28] Speaker A: Of course, more from an operational standpoint that you are aware of, are there any company that are working or already have work on solutions, software solution that institutional investors could use to build portfolios or is it still just research that we see on these topics?
[00:37:51] Speaker C: I mean there are some companies that are building these solutions but I would say they are one step away from being production ready. And I think an investor needs to ask themselves essentially are they going to be a front runner? Is this their position in the market? If they just look at other emerging technologies and if the answer is yes to that question, I think now would be a great time to actually engage with these quantum applications companies because then they have a chance to kind of shape the ecosystem. I think it's going if an investor says I want to be a laggard or fast follower, they will have to wait for a little bit because right now the, the direction of the especially the quantum application space is really co steered by of course the startups themselves, but also by the industrial players that are actually positioning themselves as front runners. I mean if you think about it, if you're a startup, if you're an entrepreneur running a startup, you of course we're listening to the, the party that actually works with you maybe pays for a POC or a pilot and then kind of shapes your roadmap. So I think this is what we are observing at the moment. So companies that have very pressing problems and are not engaging are really running the risk of not being in the focus of the development roadmaps of the applications companies.
[00:39:13] Speaker B: Lizbeth I want to build on that because I've read that there are some asset managers that are experimenting with Quantum. I'm not sure how far along they are because in our Industry, it tend to stay very dark, but it kind of raises the question and Elizabetta, this goes to one of your core competencies, which is manager due diligence. If a manager were to come to you, and let's say you're representing a family office in this case and say we're using Quantum in our workflow. Well, I mean, we all know there's a lot of AI washing today where people are saying, I'm just using AI in my investment decision making and they're using Copilot or, you know, some derivative of it that would be easy to determine. Okay, so how do you use AI, tell me the type of AI, etc. But in this case, if somebody said we're experimenting with Quantum and you know, at the end of the day your fees are going to help pay for that experiment. How do you do due diligence on a manager that says they're piloting? I'll say Quantum.
[00:40:16] Speaker A: I'm going to hire Daniel.
I mean, it seems you need, I think that you truly need a very technical expert on your team to be able to actually.
[00:40:33] Speaker B: But you'll never have that at a pension fund. At least you might eventually, but you're not going to have a Quantum, you know, technical person now. And I'm going to say many investment consulting firms probably lack the depth of knowledge to perform the necessary diligence because, I mean, what even questions do you ask? I mean, there's a few fundamental ones here about, you know, about the team, about the budget, the data and then the hardware. But I don't know. Daniel, what do you think?
[00:41:00] Speaker C: I mean, I would really look deep into the proof points of these guys. Guys. So, you know, whom are they working, what budget are they spending? Can they provide benchmarking data of before and after Quantum? But I would also say what Elisabetta said is true. It's difficult to not be fooled by somebody that is deeper in quantum, because it can be quite deceptive.
If you want to cheat, then it's definitely possible, I would say, to mask yourself as doing something that is, you know, quantum powered. And in reality it's like, you know, chatgpt or Copilot equivalent of effect on the outcome.
[00:41:40] Speaker B: So it's basically quantum washing at that point instead of just AI Washington.
[00:41:45] Speaker C: Yeah, I'm not saying that this happens at a wide extent, but it's definitely possible. So if you're not working with experts, it's definitely possible that somebody says, look, I have this PoC with quantum company X and you know, then no real output is actually affecting the outcome.
So that's definitely a threat.
[00:42:07] Speaker B: It just seems like there's real barriers to entry for a bank or an asset manager. Forgetting the asset owner side for a minute because the asset owners often are not as well resourced and we got barriers with talent, you got barriers with cost. I mean there has to be some heavy costs. And just the hardware side I assume. I mean you're not going to go out and buy something at the Apple Store and start running this.
[00:42:32] Speaker C: It depends. So if you want to buy a quantum computer that is large, it's more or less similar price text to buying an expensive HPC device. So that's not prohibitive. I would say cloud compute is accessible. I think it's more expensive than conventional cloud compute, but not orders of magnitude more expensive. So that's not going to be the problem either. I think the biggest entry barrier is again coming back to expertise because talent right now the whole software and application stack is really made from physicists for physicists.
So if you know, you don't have any physicists in your organization or you're in a niche where it's difficult for you to hire physicists, or you know, there are also not that many physicists compared to maybe computer scientists out there.
So you could end up in a situation where you are just not able to get the talent on board. So I think the quantum ecosystem has to do a really good job in improving accessibility of their solutions, making it so that you don't have to get a degree, you know, have to listen to university lectures about linear algebra to be able to use it, which could be by either making the whole software stack more accessible for non physicists or providing vertically integrated solutions that are just providing an end to end answer for the practitioners that want to run a quantum job. But I think one of these two things, and maybe both have to happen for quantum to become more accessible.
[00:44:02] Speaker A: One thing that I want to add angel here, and I think that you're very well aware of this, and we have talked about it a number of times, is the interpretability bias, I would say, you know, and I think on this side the asset ownership need to innovate also a little bit, perhaps adding at least one person, one talent on their team that is more prepared to understand this type of solutions. You know, it's for AI, it's for, and it will be more so for quantum because otherwise you risk staying on the sidelines and you know, just don't act upon the innovation that I'm guessing will come here very soon. Because sometimes, you know, the answer is, oh, I don't. You know, obviously, as Warren Buffett said, if you don't understand it, don't invest in it. And it's true. But if you don't understand it because you don't have the talent on your team to understand it, when the talent is available, it's another. It's another situation.
[00:45:09] Speaker B: Yeah, I just don't think the talent's available yet, especially at that price point. Because if someone, if there's a limited pool of talent, I would assume the asset managers could outbid the asset owners for that talent. And, you know, we're seeing phenomenal amount of money being thrown at machine learning engineers now, you know, where MEDEV offered somebody a billion dollar salary and this is in traditional AI, I mean, you know, in, in kind of quotes. So I, I don't know how an asset owner could compete. And I think what'll happen is they'll turn to an outside consulting firm and that consulting firm will specialize in Quantum when the time comes, just like they may turn to a specialized AI consulting firm to help them Due diligence on managers. Yeah, I just think the cost is prohibitive. But I think your point is you got to start thinking about this because your asset managers and your banks, I'm sure banks are using this for fraud detection and cybersecurity already, and that'll naturally flow into our space.
I got nothing else. This was great. Elizabetta, if you want to keep going.
Are you okay?
[00:46:18] Speaker A: I'm done. Thank you very much for having me. Thank you, Daniel, for your insights. Very helpful.
[00:46:23] Speaker C: Thanks, Elizabeth. Thanks, Angelo.
[00:46:26] Speaker B: Thanks for listening. Be sure to visit PI's website for outstanding content and to hear previous episodes of the show. You can also find us on PI's YouTube channel. Links are in the show Notes. If you have any questions or comments on the episode or have suggestions for future topics and guests, we'd love to hear from you. My contact information is also in the show notes and if you haven't already done so, we'd really appreciate an honor honest review on itunes. These reviews help us make sure we're delivering the content you need to be successful. To hear more insightful interviews with allocators, be sure to subscribe to the show on the podcast app of your choice. Finally, special thanks to the Northrup family for providing us with music from the Super Trio. We'll see you next time. Namaste.
[00:47:19] Speaker A: The information presented in this podcast is for educational and informational purposes only. The host, guest and their affiliated organizations are not providing investment, legal, tax, or financial advice. All opinions expressed by the host and guest are solely their own and should not be construed as investment recommendations or advice. Investment strategies discussed may not be suitable for all investors as individual circumstances vary.
