Brian Lenahan - Founder & Chair, Quantum Strategy Institute

Welcome to the Another Wisdom Podcast. We have Charles Cormier with us today. He's a founder and the chair of Quantum Strategy Institute, a TEDx speaker, the author of Quantum Boost and Quantum Excellence, Top 50 Rising Technology. And I guess you have a Substack, right? Is that this one here? Yeah. Substack is one of my favorite places to read stuff. So go follow Brian's Substack. Brian, welcome to BOD. Tell us a bit more about yourself and what you've been up to.

Well, thank you very much. So maybe surprisingly to your audience, my former career as a banker for 25 years. And, you know, I had the opportunity to retire early and then, you know, sat on the couch for 6 months and thought, what am I going to do for the rest of my life? And I found out that I couldn't sit on the couch, you know. So I thought about what I really loved during my banking career, which is you know, being the business guy on the IT projects, thought about what's going on in quanta or in the technology world back in 2016. And it was AI just getting started. And so I thought, where is AI prominent and leading? And at that point it was MIT. So went to MIT and I did their AI strategy program and the world opened up to me.

And yeah, banker, like, Canadian banker at TD. How was that? And how was TD as a company while you were there?

It was a great experience, Charles, and a particular reason because I do have a bit of the squirrel tendency. And what I mean by that is, you know, I love to do lots of different things with different challenges. And so in an organization of 80,000 people, if you have the least bit of ambition to try new things, it's a great place to do it. So I had careers in commercial banking, in operations and HR and project management and financial crimes and fraud management right across the 25 years.

Their stocks went from very low to very high. It's crazy. 160 CAD. You must be sitting on some stock. Like, are you happy about that? And why did their stock go So high. That's crazy.

Yeah. And, you know, if you dial it back 12 to 18 months, there were some challenges, particularly in the US. And so I think with a revamp of the leadership team really targeting the future, I think the market has bought into that. And that's very exciting, you know, not only as a former employer, but as a TD pensioner.

Right. What did you— why did you pivot from a banker to quantum guy? Uh, that's a pretty hard pivot. I know you sat on that couch. Did you watch quantum stuff on that couch? Like, what opened your eyes to—

yeah, no, there had to be an interim step for me, and that interim step was AI, you know. So when, uh, I was working with teams, uh, at TD, um, we were starting to use machine learning to synthesize and to analyze financial crimes and fraud management data across our portfolio. And we had real big challenges doing that because we had pillars of information that were not synced up. So it was a real challenge getting a holistic 360-degree view of a single customer. And so just getting engaged in that challenge for me really exposed me to what AI's potential was. And then when I retired, I had the opportunity to kind of think, okay, so, you know, is this something that you can pull off the shelf? And of course, back in 2016, it wasn't. It was something that was sort of brand new and really just focused on neural nets and machine learning, very different from today's world. But for me, it was an opportunity to kind of explore it and see what its future was. And so that interim step for me between banking and quantum was AI. But when I started to get invitations to some of these quantum conferences, probably around 2019, it was interesting to me. There was lots of physicists, there were lots of researchers, but there were no buyers. So to me, I was looking at this kind of saying, is this just an interesting new technology with no future and no commercialization potential? And that's why I got involved, because I thought there was white space. I thought there was an opportunity to get this from the lab into the buyer's hands. And with my experience with AI, you really needed people to be translators. You needed people to be able to say, what is this technology and how can I use it? How am I going to benefit?

And when you say AI, Toronto is home of Geoffrey Hinton, right? Like AI godfather. So AlexNet sort of happened there. So like, is that where you got the inspiration?

Yeah, I'd say there was no shortage of resources for me when I was getting started with AI back in 2016 because of Geoffrey Hinton, the teams at U of T. You know, the availability of information on the internet is nowhere near what it was today. And so going to those resources, and in fact, I started going to AI meetups right away because what I could find in those meetups was lots of people who were exploring it in lots of different ways because I knew I wasn't going to be a coder, I wasn't going to be a programmer. I wanted to be able to exploit the technology commercially. So I really got a good sense of how people were using it. And again, this is 2016, 2017. And so the community was already robust in Toronto.

Interesting. And why did you go to Quantum nowadays? Like we've seen like Claude Mitos seems like the danger is more on the AI side of things rather than quantum. Of course, when quantum hits, I'm not sure when, you can educate me on that, that will be pretty wild. But isn't most of the attention towards AI nowadays and not the quantum engine?

You know, I'd say yes, but I'd condition it. And the reason why I condition it is because, and you know, I know that, you know, over 1,100 episodes on your program, you know, you're clearly curious guy. So am I. And so when I think about what quantum's potential is, you know, for me it wasn't an urgent thing. I just wanted to understand the technology. Where the urgency comes in, as you probably know, is cryptography. You know, is it possible for a fault-tolerant, a perfect quantum computer to be in the hands of a bad actor who can use an algorithm that can break RSA-2048 or ECC, which every laptop, every cell phone uses. And so, you know, 3 years ago, that was 30 years away. 2 years ago, that was 20 years away. You know, as of the last year's work, you know, it's much more like 10 or 5 years away, but companies are not ready for that. So only the largest companies are actually putting in quantum cryptographic layers. And so really what this is, is this isn't a faster computer. This is a computer that solves very different problems. And so, you know, you take any organization, any institution who's looking to solve their hardest problems, you can try to run it on a classical computer or supercomputer. But there are problems that are just so intractable, that are just so difficult and so complex that you're going to need a quantum computer.

And when you work with a client, like, what does your mandate look like? I'd like to have clarity on that.

Today in 2026, the mandate is a roadmap. It is not necessarily an immediate investment. And so what I try to do with organizations— and Charles, they tell me all the time, they say, look, my IT budget is already spoken for. How can I possibly get involved in an investment in quantum? And my comeback to them is, number one, you need to prepare. You need to start today. But that can be as simple as putting one person on your staff or running pilots of analytical comparisons between what you do today on your classical systems and what you might do on a KISSkit or an Amazon Bracket and use that platform to compare the results. And it's exciting not only in terms of speed, but in terms of energy used. So quite often quantum is used in kilowatts as opposed to megawatts. So that can be an interesting dynamic as well.

And how do you sell something that has no relative urgency?

It was the same thing with AI. I went through this back in 2016. People said to me, look, Brian, what is it? What can it be used for? And I said to them, you need to prepare. You need to think about what parts of your business can benefit from this. And it wasn't just 2022 with the ChatGPT moment. You know, these are organizations that were looking at large databases, large amounts of data, saying, how can we synthesize and make sense of this? And so quantum is just the exact same sort of problem fast forward 10 years where the Perfect quantum computer doesn't exist, but that doesn't mean quantum technologies don't exist. Quantum sensors exist already in the marketplace today. The military is using quantum sensors on their planes to replace GPS, which is becoming jammable. So if you're in a military theater and you're flying over that area, you're using GPS. It can be jammed, but a quantum sensor cannot. And so, you know, they're already using that technology. We know it works. So the question is, how does that translate into communication? How does it translate into computing?

That's crazy, the quantum sensors. I haven't, uh, studied these. How much do these cost?

Well, this is it, because most of them are military applications, and you can imagine they're not publishing that data. But there are companies like Inflexion in Colorado who have active military contracts, um, and DARPA is certainly involved in this through the National Quantum Initiative and various others. So, you know, They are actively making investments in this technology because they know it works.

Wow, that's amazing. I'm going to definitely study that. Do you feel AGI is achievable and dependent on quantum as a hardware technology to scale?

It'll be interesting because what I think it's going to do for— you did say AGI, correct? So what I think it will do for AGI is expand the problem set that it can solve. So if AGI is the sum of all human knowledge, that's as of yesterday. You know, it doesn't mean when you— especially when I talk to clients about how you use AI, you know, we still need human beings to be creative about what's next. You know, we train AI datasets on what's happened in the past. So we really truly still need human beings who are creative, you know, but we also know that there's a huge amount of problems, large molecules, cure for cancer, you know, large city optimization that we can't solve today on any computer. And so what's it going to do is expand the problem set for AGI to be able to solve things.

Okay. And what, what kind of sets of problems can the GPU and modern computers solve? Because quantum is like 0, 1 and 0, 1, right? 0, 1 and 0, 1 at the same time. So like 3 possibilities, right? How does that infrastructure allow you to solve like, and problems, if I formulated that question well?

So one of the, one of the ways that we typically explain it is let's say you were a robot starting at the beginning of a maze where what you would do typically is follow one path until you hit a barrier and then go back to the beginning. In the case of a quantum computer, It looks at every single path at the same time in parallel. And so you can imagine the solution set is presented to you as a whole right away, as opposed to waiting for that robot to go linearly down every single route. So that's one way of looking at it. The other way of looking at it is if you are trying to optimize a financial portfolio, So you deal with a lot of venture capital guys. They are restricted to the number of factors that they can include in their calculations to, I don't know, 10, 20 in the calculation. But if they wanted to emulate reality, if they actually wanted to say we're going to use every possible factor that could possibly happen to emulate or replicate reality, that could be thousands of factors. Well, there's no way that they could do those calculations real time consistently with even a supercomputer. You can do that, however, with a quantum computer where you go from a small number of factors on a large dataset to a large number of factors on a smaller dataset.

How energy efficient are these? Because my brain goes, if it can do everything all at once, the energy consumption is gigantic, right? We need like a nuclear reactor to power it.

Or it's the opposite. It's the absolute opposite. Yeah, so remember, you're not going to use a quantum computer for— like, you'll probably continue to use classical systems for 90% of the functions that you do. So that dynamic will remain. You know, uh, the electric— electricity consumption in an AI data center continues to become— continues to be relevant. However, for the 10% that you're using a quantum technology for. That's typically measured in fractions of seconds and kilowatts. And so when you're thinking about that 10% of problems that you're trying to solve, you get an energy efficiency boost, um, just because of the way the technology works. Now, there are 10, at least 10 different modalities in quantum technology. You can have superconducting, you can have trapped ion, you can have neutral atom, you can have photonics, and they all have slightly different energy consumption patterns. Most of the people are familiar with the superconducting sort of a chandelier where you start at the top and, and the, the actual, um, structure of the quantum computer goes colder and colder and colder and colder until it's colder than outer space. And that allows you to isolate atoms within the chips so that you can start to manipulate them. And so that's superconducting. But there are others like neutral atom that are room temperature. And so all of a sudden your energy consumption wavelength is very different, like neutral atoms. So different modalities, different energy consumption, but also different costs, different scaling potential. So what I would say to you is that today in 2026, there are no winners. We're still working out the beta VHS battle in quantum, but it's also got very different signatures in terms of electricity usage.

And the fundamental difference between a normal computer and a quantum computer, what is it? Is it that different or is it like mostly fundamentally the same structure?

I think the most successful quantum computing company will, use the best of classical computing. For example, if they can make it room temperature, if they can use fiber optics, which are already, you know, existing worldwide, and they can scale using that, then all we're doing is talking about the difference between, you know, base-level quantum physics. You and I, when we're using our computers today, our laptops, we don't think about the chips that are embedded or the integrated circuits or how many there are. So most companies will never think about a quantum computer in terms of that level of hardware. They'll think about what its computing capabilities are. But when we think about the difference in terms of its compute, it's all about a certain number of quantum physics characteristics that were identified 100 years ago, and that is entanglement, where two particles are— doesn't matter how far apart they are, could be at opposite ends of the universe— if you make a change to one, there will be a change to the other pair. That's a phenomenal paradigm that quantum computing and quantum technologies is making use of. Then you've got So you've got entanglement and you've got superposition. You talked about zeros and ones. Well, in the case of quantum physics, it will be zero, one, or both until you actually physically measure it. So that's where, you know, using lasers to hold on to an atom to actually use it for calculation purposes. And I hope, I hope I'm not going too deep, but basically The other part is that they're so fragile. So if you don't have the right sort of system, they can decohere. So you lose that capability to do calculations. So there are some phenomena within quantum technologies that just aren't used in the classical space.

And if I can summarize it right, like if we take the AlexNet moment, a bunch of people thought the AI guys were crazy back then. There was a winter Quantum is sort of the same, but you had the proof at some point that this would scale, right? There's a bunch of humans still investing tons of money, energy in making quantum possible for one reason, and you probably have a proof. What is that proof, right? Like that proof of scale, because to me as a VC is like, oh, investing tons of money and energy into something that's fragile is like, mm. I've seen Rigetti Computing as well. It has not gone like that far in like decades of its existence. Right. So what's, what's the AlexNet moment for quantum?

So that's a great question. The AlexNet or the ChatGPT moment is the moment where you have a reasonable number of qubits, you know, those bits, zeros and ones, but in the quantum sphere. That are logical qubits, the ones you do the calculations on, and physical qubits which protect that logical qubit while it's doing that calculation. At the beginning, in the last 5 years, people said that was going to be, you know, over a million physical qubits. And it's very difficult to scale something that size. But in the past 2 years especially, we've had a phenomenon called error correction. And what that's allowed us to do is go from, you know, 99% accuracy, which you think is a large number, but is not when you think about calculations. If you're doing trillions of calculations and 1% of those calculations are incorrect, that's a massive, massive error rate. In the classical world, it's 99 and 13 nines behind it in terms of accuracy. So we never really think about it. So today we're at 99.99% in the quantum world. But a few years ago, we thought we would never get there. As we progress, the error correction is getting better and better and better, and the investment is now in the hundreds of billions. So this is a very different phenomenon than it was even just a couple of years ago. And for those people who are reading as much as I am, I'm seeing the iterative progress, very slow but very incremental progress towards that fault-tolerant, perfect quantum computer.

So it makes economical sense, you're saying? Like, I mean, hundreds of billions, that's a pretty large check for one company. What company would that be?

None. So I would say it's none. I would say— Yeah. So, so there are about 750 companies that are either hardware or quantum software companies in the ecosystem. The largest are around the $2 billion mark, whether that's Quantinuum or it's IonQ. So there are probably half a dozen now public, either through IPO or through SPAC that fit into that range., and the rest are, you know, sort of the— let's call it the tier 2 medium size on down. So, but that's significantly, significantly greater investment total or pool than it was even a couple of years ago.

Right. So you're saying that at some point Quantum will bring like 100x, 1,000x, 1,000,000x the performance of a computer, correct?

Absolutely. And again, for those specific—

computing is the only path.

That's right.

GPUs won't even exist in like 20, 30 years from now.

Well, so that's a good, good question. I mean, you look at functions like payroll or, you know, some factory operations, you're always going to use a classical computer because it's the right thing to do. You have large amounts of data, you have sufficiently efficient computer, and those problems are not, you know, intractable or very difficult to do. So my view is you'll have both systems running as hybrids for years to come.

Got it. Does it make economical sense for me as a VC to invest in these quantum computing companies? Because, well, I mean, it seems Initially, these computers will be very expensive, $100K each, let's say, or $1 million each. But eventually, do you feel it will be in consumer computers and the cost will fall down to normal types of computers?

I'm going to suggest— I'm going to suggest yes. And the reason is because quantum computing, at least today and probably for the foreseeable future, is a license-based system. So you're going on to a large platform using your data doing the analysis and jumping off. And we're talking about fractions of seconds. And so it's very unlikely that most people will be buying a quantum computer. The quantum hardware that's being sold is going to places like the Cleveland Clinic or Riken in Japan, you know, large organizations who are using on-prem quantum computing for very difficult problems. The consumer, even I could be using it as a consumer today where I use my dataset, go to Amazon Braket, provide it with the, with the proper quantum algorithm, run the job, and then get my result. The best thing I can do is compare that to my classical system and see what the benefit is. And then I'll do it again a month from now or 3 months from now and see the improvement in the quantum systems.

Got it. So mostly cloud, and these companies would target the Amazons, the Googles, the NVIDIA possibly of this world. Very interesting. Got me thinking a lot, Brian. Where can people find out more about you?

Sure. You can either find out— you can find me on LinkedIn at Brian-Lenehan-Innovation or the QuantumStrategyInstitute.com.

So that's Brian and Charles.