Ian Mason - Quantum and AI Tech Strategist

Our guest today is Ian Mason, a business and technology strategist, program advisor, and content developer. He has spent more than two decades as a consultant, advising senior executives on corporate strategy, boards of directors on governance issues, and product management and IT teams on the design and delivery of complex technology-based services, including quantum computing.

Ian and discuss several topics, including:

• What are the key concerns that his quantum clients express?
• Do customers sometimes fall into an 'all or nothing' trap for quantum?
• To what extent are customers fixated on the hardware capabilities?

This episode was recorded in June 2021. It is also available on Spotify, Google and several other platforms. See here

Listen to additional episodes by selecting 'podcasts' on our Insights page

‍

The full transcript is below

‍

Yuval Boger (CMO, Classiq): Hello, Ian. Thanks for joining me today.

Ian Mason:Thank you, Yuval. It's great to be here.

Yuval: So, who are you, and what do you do?

Ian:I am a tech strategist and a program advisor in the artificial intelligence and quantum space, especially where those kinds of capabilities are cloud-enabled, but I'm also a content developer. So, I help enterprises and organizations articulate their best thinking and lead industry conversations, and so on through films and infographics and animations and that sort of content. And probably not surprisingly, on the side, I'm also a photographer and filmmaker and a writer as well.

Yuval:Amazing. You're like the Michelangelo of quantum, I see.

Ian:He was better as an artist.

Yuval:How did you get into quantum?

Ian:That's a good question. I started, about a year ago, to look into cloud computing for the first time. So, that was part of a turn in my career back to technology after a gap, where I'd focused mainly on strategy and content development. And after cloud, that led me to artificial intelligence and machine learning. And the focus on cloud and AI, I think, caused me to simply trip over the IBM Qiskit cloud-enabled quantum offering. And I immediately thought, "Wow, this is a huge change." I had thought quantum to be gigantic machines that only scientists can touch and experiment with. It was suddenly being offered to all of us on the same model that Amazon and Google and so on are offering their compute services. So, I thought that was quite remarkable, and I began to dig into it, and then the science behind it started interesting me as well. And, on I went.

Yuval:So you help organizations today with their quantum strategy. How do you assist enterprises and actually, what kind of companies do you work with?

Ian:Well, this is a new offering for me. So, let's call it a prospective offering in the sense that other larger consulting firms like Accenture, for example, have quantum offerings that when you look at it, are essentially feasibility studies or conceptual, answering the question for a large enterprise, "Could this work for us once quantum reaches the right scale and the right level of the reduction of noisiness and the increase in reliability? So, could this work for us now?" I don't plan to help JP Morgan with this sort of thing. My offering is focused on smaller to mid-size businesses, but I think there are some of them out there that are probably starting to hear about quantum, beginning to think about how it might affect them or what kind of an opportunity it may create for them in terms of optimization, let's say, or how it might play into the artificial intelligence work that they're already using right now in their operations or with their customer-facing capabilities.

So, I think that's the space I'd look at, which is, how could quantum help? And if it's going to help, how should we prepare for that day? Because I think the interesting thing about the quantum industry right now, or one of many interesting things, is that the predictability of it is not as straight line as it sometimes seems when you draw it out on a map and say, as IBM does, "We're going to try and double the number of qubits or the amount of quantum volume, let's say every single year. It's not necessarily that straight line, and everyone knows that because serendipity comes into it.

And I think that companies worry in the back of their minds that it's possible that even though the roadmap, the current roadmap may say, "Oh, this might be ten years away." Or, "This might be 20 years away." What if it's three years away? What if somebody comes up with something that they were not expecting and they're not ready for that? And suddenly, their competitor's doing quantum and solving problems that they can't solve because their competitor was more optimistic about it than they were. So, I think that's really where quantum consulting comes into play.

Yuval:So, you work with companies primarily in the formation stages. They think maybe they should do quantum, or do you work with them once they put up a small quantum team to explore the opportunities?

Ian:It'd be more the second. So, not so much a quantum startup, per se, who of course has already decided that that's their space, but much more an operating enterprise, perhaps a mid-size manufacturing firm or something like that, or a mining firm or somebody concerned with logistics. And they're already using perhaps machine learning. They're getting into the internet of things. They're starting to digitize all of their operations. And they're also wondering, because they've heard about it, does this mean we should also get into quantum two years from now or five years from now? And it's that question that I want to be able to answer for them.

Yuval:So, is that the key concern? For instance, do people worry whether quantum will be useful for them, or have they answered that? Is the question more about how soon or in what area? What are the key concerns that you're hearing?

Ian:Yeah. I think one of the most common questions is simply will this be useful at all? There are two sets of businesspeople out there when it comes to quantum: there are those that fully drunk the Kool-Aid, true believers that think it's going to be completely transformative, and we have to move as fast as possible to be ready to adopt it into our business operations. And the second set, who've kind of barely heard of it. They certainly know about quantum physics for sure, but they haven't really heard of quantum computing, don't really realize that it's available over the cloud, at least as an experimental tool, and haven't answered the question yet of if this is going to be possible, A, is it going to be possible and B, will it help us in particular? And if so, how? Because, of course, the jury is still a little bit out, right?

We can run the math, and the theorists have done that. They've run the math and look far ahead and said, "If we can build X, Y, and Z, then this should give us a huge advantage in this set of problems." And that's truly amazing, hopefully, that all comes true.

But then the question is, does this set of problems apply to a given business? And if so, how? And then, is it a waiting game for them, or is it more of a, let's start to get prepared. Let's train some people on quantum. Let's run some hypothetical studies to say, "If it does work the way we expect, what kind of competitive advantage does that give us? And therefore, what's the value in it for us? What are the table stakes or the size of the prize, as we would say, that we should pursue?" and that allows them to level their investment into it. Do we build a big team? Do we just hire one person? Do we send our CTO off to do a quantum course just so they're aware of more or less how it works? These are all choices that businesses have to make.

Yuval:Do you feel that customers sometimes fall into an all-or-nothing trap where it's either quantum supremacy or nothing at all? I saw this interesting piece of research from Hyperion, I think, that says, "Hey, companies, if you can give us 5x, 10x, 20x performance improvement, that's wonderful. Even if we could hire or rent a supercomputer, a classical supercomputer that would give us the same performance." How do you see that?

Ian:Yeah, that's a good question. I think that, certainly at the level of making investment decisions in new technology, because there's such a large perceived risk around new things like this, and by investment, I even mean training up a team. It's nothing like the level of investment of should we build our own quantum computer because very few companies will actually make that decision. But even training up a team, because it's very new, you have to go quite high up the chain of command to get that sort of investment approved, which means that the question starts to become simpler and simpler as it goes up the chain of command from being a, let's say at the technical level, a highly nuanced question of, "Is 5x enough, or is it really 10x? And what if a high-performing supercomputer can do it, why are we really investing?"

These are all manager-level, technology-level questions. By the time it hits the board, where the CEO is saying to the board, "This is why we're going to put this extra amount of money into our IT budget to work on quantum." It becomes a very simple decision. It's: “look how incredibly transformative quantum will be”. That's the way that those sorts of decisions are going to get sold. So, I think there is a tendency, because quantum is brand new, to force it into an either/or bucket. But I think you're right that there will be many, many business cases where the truth of the matter is that it's actually a much more subtle decision. And in many cases, the answer may be, "For now, let's just go with the supercomputer, the classical supercomputer." But recognizing that three years from now, the quantum computer will probably be beating it, so we better also, in parallel start preparing for that.

Yuval:At a high level, what kind of advice do you give companies? I assume good advice, but if you could be a little bit more specific.

Ian:On that particular question?

Yuval:In general, on how to get started with quantum and when to enter the market, or what to ask themselves before they decide to launch a quantum team.

Ian:Well, I think, as I was saying earlier, those kind of core questions of understanding the hypothetical case of if quantum computing works as well as we all hope it does, will those situations where it's particularly good and better than what classical computing can give you, how will those situations apply to our business? For example, I would also tell a CTO that even if it doesn't apply to your business, the way your business is currently configured, you should also ask yourself how might quantum help you change your business? Does it open the door to a different kind of capability or a different kind of offering because that technology now exists, again hypothetically, a few years from now versus all of the assumptions that are baked into what you do and how you operate today, which is all based around the maximum capabilities of classical computers?

So, it isn't as simple a question as simply saying, "Given what we do today. Will quantum help?" It should also be, "Given what quantum can do, does that mean we should change our strategy or change what we offer?" especially in a competitive analysis of what everyone else is doing in your space. If no one is jumping on that and taking advantage of it, you might have an opportunity to do something quite different, given its unique capabilities.

Yuval:To what extent do you feel that companies are fixated on the hardware capabilities, number of qubits or quantum volume, and do they think about software? Do they ignore it? Do you think software is important to look at or just hardware or just software?

Ian:Yeah, I think you're pointing at an important thing in the industry that currently, the trade press themselves do focus a lot on counting qubits, which is natural. It's a competitive thing. It's easy to keep track of. And I think businesses respond to that and they see that this is one of the major factors. I think the other point here is that, because we've all been conditioned by Moore's law to believe that computing progress is defined by the number of transistors you can stuff onto a chip, and that became its self-fulfilling prophecy. Although I should be careful about that language because, of course, the effort that went into every doubling of the number oftransistors really reflected a lot of engineering work and engineering talent, and often new approaches to shrink those transistors and get them to fit on the chip with each new generation.

But that quantitative, exponential scaling down of transistor size in the classical computing space has trained all of us, including enterprise clients, to think about computing at that countable level. And I think quantum computing, because it's starting from zero, it's going the other way, right? It's actually starting, not from a macro perspective and then shrinking down; it's starting from a micro or nano perspective and then scaling itself up. But that too becomes a game of counting. And, of course, there are vastly many other factors that go into the power of a quantum computing system rather than just simply its qubits. So, I think the industry is going to be hopefully changing its language. You can see signs that it's starting to with the notions like quantum volume that IBM, at least, is talking about a lot.

And that'll lead us to a much more holistic way of thinking about why is this particular quantum computing system more powerful than this other one? What are all the factors that go into it? And that leads you to much better requirements definition and things like that, because you're not just saying, "Oh, I need a 72-qubit system." You've got a more complicated, nuanced, fully shaped idea of why do you need this system? How does it need to interface with the classical computers you're running? What kind of software do you need to be both interacting with it, and also for the kind of software that's actually running the system itself and the compilation software in between those two things, how do all those work together as a system? And I think we're starting to see those conversations and that kind of language begin in the industry, but it's only just beginning.

Yuval:And, by the way, you mentioned 72-qubit systems. One question that comes up is how do you know that you need 72-qubits? You need something that takes your algorithm or what you're trying to do and estimates how many qubits are necessary. And maybe it's not just about qubit number as the constraint, maybe you worry about accuracy, maybe you worry about the depth of the circuit or you're trying to fit into a particular hardware. So, I certainly agree with you. As we come towards the end of our conversation today, what do you think companies like Classiq could do with regards to educating the market on what's the right thing to do?

Ian:Yeah. I think education broadly, of course, is an incredibly important thing for the quantum computing industry. Again, it is coming out of its very nascent and just got invented yesterday stage. And, of course, that's highly dependent on theoretical physics, computational science, a lot of PhD level stuff, which is quite abstruse to regular people, even to other computing professionals who are used to the world of classical computing. And so, I think companies like yours, the more that they can explain both accurately, because we don't want to oversimplify the operations of quantum computing. We don't want to oversimplify the logic of how it creates quantum advantage or quantum supremacy. Because frankly, even for me, and I'm a few months into learning intensively about quantum computing, that logic of where exactly does it generate, what elements of it are, the things that generate its quantum advantage in theory have taken a while to come clear for me.

I think that even that kind of language is oversimplified in how it reaches the general press and how it reaches businesspeople. So, I think finding that level of language that is both accurate and can lead businesspeople and classical computing technologists through the learning path to key insights that you're trying to get across to them so that they see it as being logical. They don't feel too much has been left out. They don't feel like they're being talked down to. But at the same time, we haven't accelerated out of sight after the third chapter in whatever we're trying to tell them, into linear algebra and complex numbers and so on. All of which has been a temptation for the industry because that's where the magic is happening. And most people in the industry have naturally tended to need to talk at that level amongst themselves. But we need a new language for talking to everyone else, just outside the industry or just entering the industry now.

Yuval:That's excellent, Ian. How can people connect with you to learn more about the work that you do?

Ian:Well, probably the easiest way is to visit my website, which is hassardfay.com, and you'll find my service offerings and my bio, and also a link to my LinkedIn profile and you can always connect with me there.

Yuval:Perfect. Well, thanks very much for joining me today.

Ian:Thanks, Yuval.