Technology, Connected - Quantum Computing Is Not What You Think

Episode Date: December 22, 2025

Quantum computing doesn't make computers faster. It changes what's computable.Joe Fitzsimons, CEO of Horizon Quantum, explains why quantum progress is so hard to grasp: it's exponential in a way that breaks everyday intuition.Here's the math that matters:Each additional qubit doubles the difficulty of simulating the system on classical computers. Meanwhile, quantum processors are scaling faster than Moore's Law as the industry accelerates.Put those together: exponential difficulty meets exponential growth. The result is capability that quickly surpasses what any classical computer—or human intuition—can comprehend.Why this matters:Early computers didn't just speed up arithmetic. They unlocked tasks you could never complete by hand: weather prediction, aircraft design, nuclear simulation. Things that were mathematically possible but practically impossible.Quantum computing does the same—except the tasks are even more fundamental:- Drug discovery: simulating molecular interactions at quantum level- Cryptography: breaking encryption that protects the internet- Materials science: designing room-temperature superconductors- Optimization: solving logistics problems with trillions of variables- AI: training models that classical computers can't handleJoe's point: we're not making computers a bit better. We're unlocking a category of problems that were previously unsolvable—not just hard, but impossible with any amount of classical computing power.The comparison that clicks:Before computers, you could theoretically calculate pi to a million digits by hand—it would just take lifetimes. But some quantum problems aren't like that. They're not "hard with classical computers"—they're impossible, full stop. Like asking a typewriter to stream video.This short episode breaks down why quantum isn't incremental improvement. It's categorical change.If you've been following quantum computing skeptically (wondering when it'll actually matter), this episode shows you why the inflection point is closer than you think.--Other ways to connect with us:⁠Listen to every podcast⁠Follow us on ⁠Instagram⁠Follow us on ⁠X⁠Follow Mark on ⁠LinkedIn⁠Follow Jeremy on ⁠LinkedIn⁠Read our ⁠Substack⁠Email: [e-mail]

This is a five minute short on quantum computing. If by the end of it, you haven't learned something, then email me personally at mark at thinking on paper.xyZ and I'll make it up to you. Let's imagine however many years into the future, quantum computing has largely been adopted in governments, in major enterprises. What does the world look like? I think the improvements in processing efficiency you can get from quantum mechanics are so large that they are not

an increment. They don't need to incremental change. They lead to very large scale change. So if you think about the computing revolution, everything your computer does could have been done by hand. Every calculation it does. It's just adding numbers and multiplying numbers. That's

all it's doing. But not really, right? There's something, the fact that my computer can do this, but not do it, you know, every couple of seconds. Rather, it can multiply these numbers in our fraction of a nanosecond and it can be multiplying many of them in parallel, that has a very, very real consequence. I can definitely do things that I could never have done on paper, you know, not in a million years. The change is just so drastic that it enables entirely new things,

even if they are composed of tasks that could have been done in otherwise. So if you think about, if you think about quantum computing, I think one of the things that's important to understand about quantum computing is how difficult it is to simulate with conventional computers. If you have a perfect quantum computer that has no noise, no errors at all, every time you add a qubit to that quantum computer, it becomes twice as hard to simulate on a conventional computer. For every one cubit. For every one cubit.

Now, you could say, well, what rate are we adding qubits to the device? But we're not adding the qubits one at a time. instead we're doubling the size of the devices regularly. And if you look at what's happened with quantum computers that exist today, they're still limited, they still have errors, there's definitely real restrictions, they're not yet fully useful. But if you look at the rate of change, there is a kind of quantum wars law. And for a long time, the number of qubits in a device was doubling pretty regularly once every five years. And that happened from the very first quantum computers in the 90s up to, you know, 2016, 2017, 2018. But since then, as industry has started to play a part, those systems, those kind of, the largest systems up to then were mostly, were almost entirely in academia or government labs.

What has happened since then is that there has been an emergence of a quantum computing industry, where there are startups as well as established tech companies that have built up big programs in the area. that are trying to build better and better processors. And since then, that doubling time of once every five years, which is more than three times slower than Murslaw, has come right down to doubling once every eight or nine months, which is twice as fast as Murslaw. We have these systems that are doubling every eight to nine months, doubling definitely on a time scale of less than a year.

But for every cubit they add, the difficulty of simulating it with a conventional computer and the computational power of that system grows exponentially. It's doubling every time. Could you just give us an idea of what a conventional computer is in that comparison?

Because it's not my computer. Well, I mean, you can simulate a quantum computer on your computer, but the size of the quantum computer you can simulate on your computer is smaller than the size of the quantum computer you can simulate on a large supercomputer. computer. So you can probably simulate, you know, 28, 30 cubits on your computer. You know, you can probably simulate 40 something on a large supercomputer. But once you say 50, 60, 70, it's

impossible. 70 perfect cubits for a deep computation, a long computation, you've no chance of doing that. Like that requires so much memory. It's just not something you can conceivably do. And the largest quantum computers today are not at 70 cubits, there are 1,100 cubits. And that is very far beyond it. Now, there's still errors that occur in these, and those errors make it easier to simulate. But if you were able to squeeze out those errors and you have 1100 cubits, there is no conventional computer on Earth that could come close to simulating that system. Now, does it mean it's useful yet? Not necessarily, because you need to be able to take a problem of interest and fit it into a device that small.

You know, with that, 1,100 qubits is still quite limited because if we think of a computer with 1100 bits, there is a computer with about 1,100 bits, and that's an Atari 2,600. And it's where we are. Yeah, exactly. That's pretty crazy to think about. Does simulating quantum work on conventional computers, Is that the process we have to go through to figure out the usefulness of them just because we can't,

just because our intuition isn't there to do it? Talk to me about the simulation piece of this. No, not at all, actually. I think it's quite common with beginners that they like to use simulators to be able to get very quick feedback to try different things and see how they work. But actually, the way we understand these algorithms, the way we prove they work and so on, tends to be more pen-of-paper mathematics. So we tend to write down in general description of what it's doing

and from that, you know, figure out how many operations it would require and so on. And understanding how to take advantage of quantum mechanics, a lot of the time you care about doing transformations of the data. We call these quantum Fourier transforms, but you're taking a Fourier transform of your data again and again and trying to use that in a constructive way. and that's a challenge, but it's something that we understand the mathematics of pretty well. Now, how to do useful things, that's more challenging, but writing down the net effect as a mathematical object, yeah, we can do that fairly well.

But the place I was going with this is basically to say that combination of the quantum computers growing exponentially in size with the exponential difficulty of simulating them, means that the power of quantum computers, or the power of quantum computers, even compared to the conventional computing as it's growing exponentially at time, it's double exponential. And this is not something we're used to. There's very, very, very few things in the world that grow at a double exponential rate. But if you look at quantum algorithms for machine learning, for example, I used to work on some of these before I started Horizon. And one that we had worked on was for a particular machine learning model called Gaussian process regression. but if you wanted to train a model with a thousand

training points and then you want to get a prediction out of that that has some cost on a conventional computer is going to take some number of operations but if you were then to increase to a million data points the slowdown would be a factor of about a billion

on a quantum computer the slowdown is a factor of four and if you were to go from a million training points to a trillion training points, the slowdown would be another factor of four. So when we think about what quantum computers can do, it's very hard for us to comprehend in terms of some of these things, the speed up, the change with time is so enormous

that it's really hard to comprehend. And that's why I say it's like that change from not having computers to having computers, from when computers were people to when computers were machines. And I think it's going to be that scale. change. Now, if you ask others, they will give you different answers. So I'm not going to pretend that my view is shared by everyone else in the field. For sure, it's not. Some people view them as just, you know, devices that might be useful for solving optimization problems and things like

this, but just some kind of special purpose thing. But I mean, I wouldn't be in this, so I wouldn't be doing this if I wasn't optimistic about it. At least from where I stand the evidence seems to support it at the moment that we are on this path that can potentially lead to to another computing revolution, like a second revolution in computing that is at least as big as the first.