The team discussed the potential and challenges of quantum computing, with a focus on its applications in various industries and its potential to revolutionize fields like cryptography, optimization, and drug discovery. They also explored the process of adopting new technology, the importance of educating oneself on quantum computing, and the need for C-suite buy-in. The conversation ended with a discussion on the potential of quantum computing in solving large-scale problems, the availability of quantum computers, and the potential applications of quantum computing in various industries.

[00:00:06] Speaker A: Greetings everyone. Welcome to this week’s coffee with digital trailblazers.

Interesting week of weather across the United States at least, especially for those of you in the Southeast. I’m really excited that you’re here.

We’re going to give our normal few minutes for everybody to join this topic that I’m super interested in.

I am nowhere near an expert. I invited a couple of experts to join us today, but we’re going to be talking about quantum computing and what digital leaders in particular should know about this technology.

We will certainly get into where it is on the horizon from science to engineering to application.

We will talk about some of the applications, some of the concerns around this technology.

And I am just super excited to learn as much as all of you are. We always say how important it is to be lifelong learners.

And you know, whenever there’s a new technology, we’re not quite sure when it will be ready for early adoption.

When should government and large enterprise really be investing time and energy into it?

When should the rest of us start getting more knowledgeable about it? And so we’re going to try to help all of you get onto that spectrum. This week we are going to use a common stream again. Please do say hello if you are here.

Super excited to see a bunch of people using it last week. So say hello there. I will be pasting in the questions there as they come up and I’m going to share some links with you. I did some research this week just to get slightly more knowledgeable today I have a link from the Wired Guide to Quantum Computing.

This is one of the Better 101 articles. This one I would say is a little bit more skeptical. There’s a quote I have in here. If you squint out the window on a flight to San Francisco right now you’ll see a haze of quantum hype in Silicon Valley. But the enormous potential of quantum computing is undeniable. I shared a link from Pascal.

Two of their leaders are here joining us today. They have an article on the essential guide for business leaders ready to innovate with quantum. So there’s an article I’ll share around that. I’ll talk about Google’s unveils unmind boggling quantum computing chip and what that chip is doing and how that’s potentially a game changer.

We’ll talk about the risk in terms of cryptocurrency, crypto, cryptography, sorry, and where that is. I have Some links that $49 billion in global quantum investments, that’s I believe in 2024 was the projection estimated to get to 200 billion by 2030?

And then I have a link for practical applications on quantum. But without any more delay, I want to introduce Michael Warren. Michael, thank you for joining us.

We’re just going to start with our first question.

We do a lot of discussions here, Michael, where we de jargon a new technology or a buzzword for our audience.

So I’m really interested in your opinion.

What should transformation and technology leaders understand about quantum computing? Hello, Michael, Good morning.

[00:03:47] Speaker B: Good morning, everyone. Thank you for inviting me.

Joining me on the call is Christian Beno. Christian is a senior technical advisor for Pascal with years of experience in this area.

So my background briefly is my eighth year in quantum computing, all on the commercial side. I spent 10 years, 12 years rather, at intel doing global innovation projects. So you can read into that as meaningful, introducing new and novel technologies that will drive innovation and ultimately drive earnings per share impact. So going way back and dating myself, you can think of that as virtualization, cloud computing, big data, software defined networking.

You have AI. Quantum is certainly one of those things.

So what I’d like to do and just give a quantum 101 and actually be about 2 minutes long is understanding what quantum computing is, how it’s different than traditional computing, really. Quantum computing is a revolutionary approach to computing that uses the principles of quantum mechanics to process information in fundamentally different ways than classical computers.

While traditional computers store and manipulate data in binary units called bits, either ones or zeros, quantum computers use quantum bits or what we call qubits. Qubits have the unique ability to exist in multiple states simultaneously, thanks to quantum phenomenon like superposition and entanglement.

Albert Einstein described this as spooky science, which it you can certainly look at it that way.

But in superposition, a qubit can represent both a 1 and 0 at the same time, vastly increasing the number of possible states a quantum computer can explore.

Entanglement, on the other hand, is when a qubit becomes linked such that the state of one qubit can instantaneously affect the state of another, no matter how far apart they are. Now, that is a difficult concept to get your head around, but absolutely that is the case.

These quantum effects enable quantum computers to perform complex calculations far faster than classical computers, particularly for certain types of problems like factoring large numbers or simulating molecular structures.

While quantum computers are still in the early stages of development, there is a robust ecosystem out there with no less than a half dozen different vendors having machines with different modalities. They hold the potential to revolutionize fields like cryptography optimization. So you can think supply chain optimization or financial optimization for derivative securities, drug discoveries.

However, building practical large scale quantum computers is a massive technical challenge due to issues like qubit stability or what’s called in the science quantum decoherence or error correction. But progress is being made.

Many researchers believe quantum will eventually offer capabilities far beyond what’s possible with today’s computers.

So a question I get all the time is, how do I get to be quantum ready? What does that mean?

And actually it’s no different than what your companies went through adopting any new technology. So let’s say that’s cloud computing.

Actually, it’s about a four step process. You need to learn, connect, you need to access, you need to get buy in. So the first thing when I meet with companies and I find out they’re just dipping their toe in the water for the first time is, you know, you have to go out and educate yourself on the technology. And there are a number of ways to do that. There are plenty of online courses.

My old company, Q Control, has a very good quantum learning process. It’s one of the better ones on the market.

That is for fee, but however, there are a lot of free resources out there that you can become smart on quantum computing and become conversant.

The second thing that really comes up, and I found this to be true, no matter how big your company is, if you’re in the Fortune 500 realm, you have scientists at your company that have been kicking around in Quantum for a number of years, but you don’t know who they are.

I have seen this firsthand, a very funny story of being at a company in Germany, actually standing in line at a commissary for lunch. And we were continuing our meeting and we start talking about Quantum again. And a voice behind us said, oh, I’ve been working on Quantum for 10 years now. We all kind of looked around and our host said, well, welcome to the Quantum team.

My point in this is there are scientists out there working for your company that are in the shadows and you need to bring them out. And you can do that by creating a special interest group at your company, sponsoring just a lunch and learn.

And you will quickly find out who at your company has been looking at this as a sort of a quantum hobbyist.

As things progress from there, a next logical step would be to identify those intractable use cases and when, I mean intractable problems that you cannot solve classically.

A good example would be the traveling salesman scenario.

If you went out to your line of business managers and had A conversation around what problem that you can’t solve now, but if solvable would yield a either high ROI or even an earnings per share impact to your company.

And if you kind of frame the question like that, I think you’ll find a lot of line of business managers do have these types of problems

[00:10:37] Speaker C: that

[00:10:37] Speaker B: they would be willing to talk about with you.

And sort of the last piece of becoming Quantum ready, you know, after you educate yourself and draw people out of the shadows, identify your use cases, the last step is really, really critical and it’s, it’s getting the buy in of the C suite.

So educating the CEO, educating the CFO or the chief operating officer as to the benefits of Quantum and actually being very upfront with the C suite. The ROI in Quantum is not measured in a week or a quarter or two quarters out.

AI, you can do that now, you can get immediate results.

This is going to be a long term play.

And what we’re finding, there’s no one industry that is really leading the way, but rather every single vertical out there, industry vertical, there are two or three leaders, trendsetters that are looking at that. So that could be big banks, big insurance companies, logistics companies, healthcare and the such.

So I know I went longer than two minutes there, but I hope that sets for everybody.

[00:12:00] Speaker A: Thank you. Michael, Christian, welcome to the floor.

I’m wondering if you can continue down with this and maybe dive into the time frames, the time horizon, Michael, saying we should learn about it and form our special interest groups for those of us who have reached into our executive committees. We should get them more knowledgeable about the opportunities.

But you know, I’ve heard everything from three to 30 years and if it’s three theaters I care about it and if it’s 10 to 30 years, I probably don’t care about it. So your perspective, Pascal’s perspective on the timeframe where large enterprise and government need to be stepping up and putting some resources against the opportunity around quantum computing.

[00:12:50] Speaker B: Yeah. So Christian, why don’t I take first swag at this and then I’ll let you jump in the. I think you all saw the announcement by Nvidia’s CEO last week saying, you know, we’re a long ways away, 15 to 30 years out from a universal quantum computer. And I think he was referring to a fault tolerant quantum computer and that’s probably pretty true.

That’s way out on the horizon. Interesting enough, Nvidia is partnering with every single company in the Quantum industry.

So it’s sort of interesting to hear that. But what we have now are computers which are in the NISQ era, which is the near term or near term intermediate scale devices.

And these are not fault tolerant, these have errors, there’s a large amount of error correction going on.

There are several different modalities from superconducting qubits to neutral atoms, which Pascal does, to ion trap to photonics, you name it. And we’re all kind of hustling every single day towards providing that. And we are seeing small incremental step ups in reliability and efficiency.

But Christiana, I’ll let you chime in from here.

[00:14:16] Speaker C: Yes, absolutely. Hello, Ariana.

So it’s, I would say it’s a multi layered question asking about timelines because there are multiple things to consider.

So indeed, as Michael has evoked, there is this distant horizon for reaching close to perfect or perfect ideal universal quantum computer, which is really the ultimate goal and holy grail for all of the currently existing hardware providers.

But that’s not the scale where quantum computing will become useful.

So for this to illustrate, I think one key information is that 2025 has officially been declared the year of Quantum by the United Nations.

The initiative behind this is really to bring around global awareness for this new emerging technology, especially due to the fact that it is of quite disruptive nature for quite a few different industries and types of problems that Michael has evoked. A couple of them.

What we are expecting in the next few years is that the current generation of quantum computers, so what we refer to as this NISK noisy intermediate scale quantum computers can already bring value to certain specific, well targeted use cases.

So this value can manifest in very many different forms. It is a function of your key performance indicators, your performance metrics, which can be dependent on the problem you are tackling as well as on your industry.

But this added value, this not necessarily leads to an advantage. What typically, typically people tend to refer to this so called quantum advantage. We try to be very prudent with this word.

We are not there yet. But that doesn’t mean that you cannot improve the existing processes. You cannot make the exploitation of your available computational resources better, more efficient by introducing this new way of computing into existing workflows. So it is an overall learning experience not just for us, but but for the end users as well. And the collaboration is key in better understanding. Therein lies the impact in a short term, because there can be impact in a short term in the coming years.

One famous example where we are expecting the most impact in the near future is material science related tasks which is mostly still on the R and D divisions from physics perspectives, but still can have a cumulative impact for a longer term for the materials we deploy, for coating, for paints, for constructions, etc. Etc.

There is also this longer timeline that a lot of the companies are aiming for as first main objective which is situated between 2028-2030 reaching fault torrent quantum computing, meaning the practical introduction of logical qubits instead of physical qubits. So switching to logical information units instead of the physical ones currently in the machine in order to reduce drastically error rates to approach better to what we have for instance in your everyday currently existing classical computer. This will enable much more diverse applications as well as new ways of handling information information, be it from a machine learning perspective or a simulation perspective.

And one final comment about your skepticism for looking at more long term, I think we will going to talk a little bit about the cyber security aspect.

Ever since the 90s, actually 1994, Peter Shor, mathematician, has devised an algorithm that is able to factor numbers much more efficiently than what we are capable of doing currently with classical computers by the use of a powerful enough quantum computer.

The key message with this algorithm, Shor’s algorithm, is that this is a risk for almost everyone because most of our current data sharing and cryptography processes relies on the fact that classically we are unable to perform this task, this factorization of integer numbers in an efficient way with the available computational resources.

So under this assumption, a lot of the cryptography keys currently existing are operational and quite robust.

However, the data that you’re sending now is not just for current usage. This is something that you want to keep secured potentially for 10, for 15, for 20 years. And herein lies one of the main risks and main points of interest is that okay, maybe a quantum computer cannot yet break existing cryptographic keys, but if in 20 years it will be available, people can already acquire encrypted data. Today they can keep it for 20 years.

Once the encryption method is operational, they can decrypt it.

So in terms of thinking about risks and potential impact, the time horizon needs to keep this also in account.

[00:20:23] Speaker A: Thank you Christian.

Before Joanne, we go to you, I want to get your opinion on this.

I’m hearing kind of a horizon of timeframes, but there’s a great common stream going on on LinkedIn. And I want to thank John, who’s shared a resource there that I’ve captured. He’s talked about an event in New York City for quantum computing and financial services.

Navid has shared a link for the Quantum Innovation Summit in Dubai.

So more information is Coming out on the common stream. This is obviously an area that we’re all sort of trying to learn from each other from. And I just point to you to go to that resource for that. Joanne, short term, long term, where’s your mindset around this? And is there an aspect of quantum computing that technology leaders and transformation leaders should know about that we haven’t covered yet?

[00:21:21] Speaker D: Okay, first of all, I’m in the NIST era too, the noisy intermediate scale.

I see a lot of work particularly coming out of universities that are aiming at commercialization potential of what they’re doing with quantum around materials, photovoltaic. Think about a car that’s painted with a new resin or a new capability that’s a liquefaction of something that allows it to generate its own energy as it’s driving.

Right? We see some of this in things like asphalt that’s used in certain countries that actually generates a capability for an ev.

So think about now, an EV that’s painted with a substance that allows it to capture sunlight and turn it into energy.

So there’s a lot of work in quantum computing around that. Things like materials, also drug discovery in the short term. So my time when I’m much more optimistic, and maybe not from a naive perspective, I would say, but optimistic that these niche areas are going to start to come to fruition much faster than people anticipate because the universities are already there in certain instances, niche problems. But they’re very complex niche problems that need to be solved. In drug discovery, it’s new chemical entities that perhaps will bring cancer drugs to market much faster than the 10 or 15 years they currently take.

So I see a timeline of in the next three years we are going to see companies not only looking to commercialize quantum computing, but that the results are going to be so dramatic that they will absolutely take the world by storm. Much in the same way as, you know, OpenAI did with, with LLMs.

We’re at an inflection point. And that inflection point is like crossing a chasm. We’re just getting to the other side. You know, I mean, there’s a lot of work that’s being done through the Quantum Economic Development Consortium. There’s defense stuff. There are, you know, Rigetti, IonQ, D Wave, all of these are specialized quantum startups. So we’re going to see them. And I think leaders have to be aware that if you have giant supply chain problems, if you have complexity in your engineering in particular, or in material sampling or materials development, that’s where you’re Going to need to be prepared very quickly. And I guess I would say as well, you know, it’s not just the risk of cryptography. There is a talent shortage around this.

These are problems that are not only going to be addressed in terms of very short timescales, but we need to start training people how to use Quantum. It’s a whole different development mindset than we’re used to.

And I think you have to start thinking about how you’re going to actually make this work in your own company.

[00:24:40] Speaker A: Thank you, Joanne. So Joanne’s sort of like a balanced approach and looking for opportunity, near term opportunities in some concrete areas. Martin, thank you for joining. I see your hand raised. I think you have a really good question that I see here in the comments. So go for it.

Well, I got first of all a statement.

[00:24:58] Speaker E: I think one practical opportunity of Quantum, but that obviously depends on quantum computing power being available in the mainstream is permutations and combinations. Anything that involves massive permutations and combinations and solving that, such as dynamic supply chains where you have the opportunities for routing and everything else. I think those types of problems are as well as things like material science and new drugs and things like that. Those types of problems are also very suited to Quantum, which is why the cyber security and cracking cyber security because it’s permutations and combinations.

But I had a really good question for Michael which is how do you explain this to the C suite? The C suite, normally fairly reticent. They’re reticent about spending money, whatever else. How do you sell this to the C Suite?

[00:25:50] Speaker B: That is a great question. And I’ll tell you what not to do. First, do not sit in front of the CEO, COO or CFO and start talking about quadratic unconstrained binary optimization, for instance.

Their eyes will glaze over and you’ll lose their interest.

[00:26:09] Speaker A: Michael, my eyes are glazing over.

[00:26:15] Speaker B: Yeah, not being a scientist, I was very proud of myself learning that acronym by the way.

But you know, seriously, the conversations you have with the C suite are, you know, concentrate on the ROI of solving those intractable problems. And, and I’ll go back to a conversation I had with a CFO at a company who had a multi billion dollar supply chain and we asked what would it mean to your company if we optimize your supply chain by 3/4 of 1% to 1% year over year. And his eyes lit up and he said, well, I would be the next CEO of my company if I could do that.

So outlining a intractable problem and then Doing your homework and finding out if solvable, what would that mean? Because C suites, they’re concerned with earnings per share. They’re, you know, if I invest money, when am I getting it back and what could be the possible reward for that? So I have always found and been successful in my career in Quantum over the last last eight years of having those types of conversations and kind of pushing the science to the side.

I hope that answers your question.

[00:27:34] Speaker E: Thank you, Michael.

[00:27:36] Speaker A: I think we have another question from John or a comic. Hi, John. Welcome to the floor.

[00:27:40] Speaker F: Hey, good morning. Yeah, thank you for having me here. So Warren and Christian, I was at IBM in 2016 when they kind of had their press release and then they had that version that you could play with on the web of their quantum computer. But, but how real are these things today? Like, what are, what do companies have? Like, what are they doing? Like, I would love to hear, because I haven’t really heard anything about them for, for eight years. And so, like, what do we have today?

[00:28:05] Speaker A: Yeah, that’s, I don’t know who could comment that. It’s very similar to my question about, like, if I had access to one of these things and I was a developer and, you know, maybe I know some things about this, what does the experience look like today in throwing a problem at a quantum computer?

[00:28:24] Speaker B: Okay, I’ll take a stab at that and we’ll let Christiane add the technical aspects of it.

So right now you’re looking at a couple of major buckets for quantum computing.

There’s the chemistry and material science for pharmaceutical drug discoveries, or companies like BASF or Dow Chemical.

Optimization is another big bucket around supply chain and financial optimization.

Machine learning is another one crossing over into AI and large language models. So each company will have very specific use cases. And it all deals with developing the algorithm to address that. There are some algorithms out there that will certainly further the cause. And Christiana, I’ll let you take over from here.

[00:29:20] Speaker C: Yes, definitely. So actually with Quantum in particular, even more so than other emerging technologies, the very nice thing is that it’s still very much academia and science driven.

And a particular consequence of this is that the community itself is very open, open and sharing results, open and sharing resources, open and sharing materials. So actually online you can find a lot of introductory materials and a lot of highly technical materials.

The problem is really trying to go through which one is more adapted to what type of audience.

But in terms of resources available, resources, already everyone can pick their preferences.

What we have noticed as a trend is that hyperscalers Crowd providers are also positioning in terms of providing access to quantum computers and building this as part of their programs and offerings for people to be able to test it out, to get a feel for it. Because that’s the first thing that you need to establish when you are trying to get into the technologies. Is it actually useful for me? How can I make it useful for me?

And part of the answer is really as it was already mentioned by John as well, is trying to tackle the right problems.

But part of it is also getting a better idea, better understanding of it. And for that there are some very, very nice tools and resources already there and available for anyone to use them.

[00:31:17] Speaker A: So what does a tool look like? Christian? If I, if I were trying to do an experiment and had access to something, what would it look like? Would it. I don’t, I’m not, I’m picturing an ide, but I don’t think it’s an ide, is there?

[00:31:33] Speaker C: So the tools, at least the tools I know of are from a programming perspective are very much Python based.

So from a developer, like a programming developer aspect, the Python based libraries, but still open source dedicated libraries are predominant in the industry and the level of abstraction and how high level you can get from these basic elements is sort of dependent on the technology, on the company, on the provider.

You can get access to interactive jupyter notebooks, you can get access to graphics user interfaces that work in a no code environment.

If I may cite for instance there is IBM Qiskit composer that allows you to just play around with quantum logic gates and literally without any coding you can just move around floating virtual boxes or on Pascal side we have Pulsar Studio which allows you to play around with small atomic systems. All of these tools run on your web browser with your web browsers available resources so they are quite user friendly in this sense.

[00:32:53] Speaker A: Thank you.

[00:32:53] Speaker C: You can also find some more much more deeper and in depth resources. It really depends on your level of maturity and how at ease you are with either the mathematics, the physics or the informatics side of it.

[00:33:12] Speaker A: Thank you Christian. I’m going to go to Joanne next. Joanne. Let me take my quick break folks who have joined us for the first time you are at the weekly coffee with digital trailblazers. We meet on LinkedIn every week to speak about topics for digital transformation leaders, often leadership practice, governance. And today we’re talking about emerging technology with quantum computing and what digital leaders need to know. I want to thank again my special speakers Michael Warren and Christian Benio from Pascal who are here to explain to us what quantum computing is all about.

I’ve done some reshuffling of the updating calendar. Next week is Data Privacy week and so I decided to switch around the calendar a little bit. Next week we’ll talk about how to take control of your data, which is the Data Privacy week theme this year. And I’m hoping to have a couple special guests for that one.

On the 7th we’re going to talk about establishing product management in non tech industries. This has come up a few times, but in terms of building up your skill sets as digital leaders, most organizations have agile going on, have design thinking going on, but product management is still a work in progress. I’ll talk about on the 7th. On the 14th we’ll talk about workplace transformation in the AI era, embracing new roles in skills. And then on the 21st we’ll be talking about building smarter organizations, transforming to intelligent operations. So those are the four that are upcoming and as you all know, use the URL starcio.com Coffee Next event. You can see it in the top right hand side of your screen. That will always Redirect to the LinkedIn page where we’re hosting our next event. I’ll switch that over later today. So Joanne, before I get to my next question, I think you either have a comment or a question for the group.

[00:35:14] Speaker D: It’s, it’s both actually.

The question is for Michael, would you agree perhaps that the way to get to the C suite in terms of early education and then later investment in Quantum is a value based argument. I’m, I’m hearing this over and over again from those people who are very early adopters, let’s say, and, and have commercialized capability out of university research.

That their argument, their sales perspective is we, this will add tremendous value not to your top line or your bottom line, but to both and your overall resiliency as a company.

Do you agree with that?

[00:35:59] Speaker B: Yeah, absolutely.

Yeah, I absolutely agree with that and that’s an excellent point.

We’re seeing a lot of collaboration between Fortune 500 companies, government and the university sector on jointly developing solutions. But your point about value based selling to the C suite is absolutely spot on. That is definitely the way to go and that’s something that resonates, resonates with them and that will certainly help create, buy in for any Quantum project at organization

[00:36:34] Speaker D: because I’m looking at this from the point of view of training. Also like somebody asked me not that long ago, maybe a month and a half, you know, they’re starting to hear resonance about Quantum and there this is a company in the life sciences industry and they’re looking at how do they begin to retrain, not augment the training of, but retrain some of their IT folk to gear up to use Quantum. And that to me is not only a value based argument to the C suite, but it’s also value add to the individuals. Like, you know, we’re seeing a lot of companies having to retrain or augment the learning of their staff for AI.

To me this is akin to the same thing.

[00:37:25] Speaker B: Yes, I certainly agree with that as well.

If you look at, well, stick to the pharmaceutical industry. If you have a computational chemist that is, you know, doing drug discovery work, for instance, and he or she has the interest in Quantum, continuing education would certainly make them more valuable than organization.

A number of years ago I would, I had chemists working for me when I was at Zapata Computing in Boston. We were working on several projects and to upskill them, I sent them to Will Oliver’s course at MIT.

And that’s specifically designed for PhDs.

Very, very heavy on the math, very heavy on linear algebra. But that is certainly a great way to upskill your existing workforce and create value within your enterprise.

[00:38:21] Speaker A: Yeah, if I’m going to chime in here, Michael, I like this, you brought up linear algebra and that when I finally made the connection over what types of large scale problems Quantum is going to be good at, that was sort of the first skill set that I was thinking about. I was thinking about, you know, all that, all that algorithm that goes into mapping technology and how do you plot a course across and optimize it for certain variables and how hard it is to think about the programming around that, because there’s a bunch of different paths that you can go down. But if I think about programming that linearly, I need to go doing a lot of sequencing to get down to an optimal path. And I think this is a big part of what Quantum is trying to solve for, is using the probabilistic nature of cubits to be able to do that more efficiently. Do I have that right?

[00:39:18] Speaker D: I would say yes, because if you look at what’s going on with agentic AI and also symbology being used to do, instead of writing a long, long, long, long prompt with many variables, there are math oriented and math specific large language models that could be used. I know that sounds antithetic, but the notion of symbology is to convert the very long natural language prompt with all the variables into algebra.

And it’s amazing the difference that it makes and I think that might be the intermediate step between those working on AI or aspiring to be digital trailblazers in quantum. That that would be an area of focus and I’m not sure if that’s correct from, from Kristen’s point of view or from Michael’s point of view, but that’s how I see the world.

[00:40:13] Speaker A: Let’s go to Joe. Joe, I think you have a question here. I see in the comments that I think is a good sort of variable to bring into this discussion. Yes. We haven’t talked yet about the potential

[00:40:25] Speaker C: cost of using this technology.

[00:40:29] Speaker B: At least in my limited vision of

[00:40:33] Speaker A: what these things are.

[00:40:34] Speaker B: You know, there are these multi billion

[00:40:36] Speaker A: dollar installations with a priesthood of technicians around it and I just wonder about the cost structure.

What does it cost to use these things?

Go ahead, Michael.

[00:40:49] Speaker B: Yeah, I can take, I’m sorry, I was on mute there. I’ll take a run at this.

There are two ways to access quantum machines right now in the NIST care and that is, is you know, over the, over the cloud. With you know, you can do that with Pascal, you can do that with IBM, you can do that with a host of others. And the other is actually having a machine on Prem, which we’ve seen mostly governments around the world. Pascal has a number of on PREM machines that are sitting in data centers right now. And you know that can be you know, 20, 20 million euros and up up. So they’re, they’re very expensive.

But it all really comes down to the algorithm. In the use case that you are trying to run, it’ll dictate what your costs are. Christian, what’s been your, your experience in that area?

[00:41:44] Speaker C: So I would say that it is, yeah, the cost is, is one thing but with that this is something that’s as I mentioned, still scarce resource.

So it is being off balanced by that fact, the current cost schemes.

But on top of that there are many alternatives.

And at the whole, I would say development process itself when you are designing quantum solutions is tailored for the fact that you do not immediately need your the access for a quantum computer. Indeed, when you’re developing methods and you’re developing solutions, a lot of the implementation and a lot of the development task is happening via what we call emulators.

So you use conventional classical computational resources like CPUs or GPUs in order to simulate the behavior of a quantum computer or emulate the behavior of a quantum computer.

You can use classical resources that are available in a very, very inexpensive way in order to get a first feeling of how problem Formulation problem implementation works in a quantum framework without actually having to rely on quantum computer. Of course if you use classical computational resources for this, you will never be able to actually showcase what is the added value, what is the advantage of it. However, just getting a better idea of it, drawing some conclusions, reaching initial consensus of what I can expect if I were to rely on quantum computers in this or that specific task is already helpful to get an idea of.

Does it worth to go to the next step and actually rent out a quantum computer or eventually to buy an actual quantum computer for the usage itself?

I hope that Replies to a question

[00:44:08] Speaker A: I know Joe has a follow up question, but when you talk about 20 million euro machines, are these actual quantum computers that I can buy and put in a data center?

I didn’t think that was available just yet, but can you clarify what that machine can do?

Is it a simulator or is it an actual quantum computer?

[00:44:29] Speaker C: Michael Christian it is the machine. So in Pascal’s case we have our first generation analog type neutral atoms quantum computer available for purchase and it is a machine that has already been deployed here in Europe where I’m based in two high performance computing sites.

One is currently being subjected to its final operational tests before putting it on the actual run as part of the French Atomic Agency site and the second machine has recently been installed in the ULI supercomputing center in Germany. So these are analog quantum computers.

So not just an emulator, not a simulator in this sense, but actual Quantum computers with 100 physical Qubit available at their disposal.

[00:45:31] Speaker A: And what does the word analog mean in this case?

[00:45:34] Speaker C: So analog in this case means that the computation process does not rely on quantum logic gates.

So it is not what we typically call a digital quantum computer. The computation procedure is happening in a continuous in time fashion.

In our case it is with laser impulses. Depending on the technology, it’s done with different methods. But what it means being analog is that you do not have discretized sets of instructions that you you transfer to your units of information to carry out the computation, but you do a single continuous in time evolution of the information evolution of the quantum system in this case.

So it is akin to what has existed in the 70s.

So there was something called an analog computer that worked with specific impulse signals and that also relied on this continuous in time manipulation to carry out computations.

However, in a classical computational framework, given that data itself is binary, it is something discrete, it is digital in this sense, using quantum using logic gates for a classical computer made much more sense. Therefore this analog way of computing in a classical computer has died out since then.

However, since for quantum computers information is much more complex, as it was highlighted in the beginning, it’s not just a zero and a one. It can be, in a sense, everything in between.

Information itself is continuous, so why not make the computation process itself a continuous action?

[00:47:23] Speaker A: I’m just wowed. I didn’t know we were at that stage yet. This is not an area I follow much. And Joe, I think you have a follow up question. Martin is raising his hand. Go ahead, Joe.

[00:47:34] Speaker C: Well, you got into it.

My question really was answered quite extensively there.

[00:47:42] Speaker A: I was also curious about whether I could go and buy a quantum computer next week. But it’s been answered, so. Thank you, Isaac. Thanks Joe. Martin.

[00:47:52] Speaker E: So I’ve got two questions.

First one is to what extent is the lack of off the shelf applications to run on a quantum computer holding things back? You know, I kind of almost align it to when we were learning C back in the 80s and we didn’t have lots of libraries to call on.

[00:48:12] Speaker B: Almost, yeah, that’s a, that’s an excellent point. And if you leave scientists to their own devices, they will do science all day long. But the adoption of any technology really depends on developing those applications for very specific use cases that solve problems.

And that’s where we are right now. And that, that’s really something that Pascal is doing is kind of taking this out of the spooky science world into creating different applications for, you know, credit risk scoring or optimization of electrical grids.

But that’s a good point. Until you have applications that solve problems, it’s just a science experiment.

[00:48:59] Speaker A: So, so let’s follow up on that. I had a question here. What are the applications that, you know, you mentioned material science before, financial services, insurance, you know, those that should have people looking into these problems. What are some of the early problems that you see people working on right now?

[00:49:22] Speaker B: Some of the early problems you can sort of go by, you know, we’ll stick with pharmaceutical, you know, everyone’s interested in protein folding. Well, that’s an extremely complex problem and it’s not going to be solved for years even with a quantum computer right now in its current state. But what people are doing is they’re looking at, if I can, I’ll butcher the science by saying kind of less intractable if you will, not, not trying to hit the grand slam, but pick problems that maybe don’t have a lot of variables that are solvable in the near term. And then as the technology gets better and the hardware gets better, certainly the algorithms need to get better as well. So it’s sort of a whole ecosystem of quantum from the hardware, the software use cases and talent all coming together to drive towards a common goal. But Christian, what has been your experience there, especially on the algorithmic side?

[00:50:25] Speaker C: So recently what we have noticed is an increase in interest from especially transportation and mobility industry as well.

And this also ties back to some of their increased needs in terms of either logistics, scheduling or in general optimization type tasks that lend well for a quantum computing based approach to propose solutions.

So I would say the question is not just what is the right industry or which industry can benefit from it most, because all industries can find the specific parts of them that can have lasting impact for this new intelligence, for this new technology to be included.

If I may mention, a particular example that we have worked on recently is coming from the energy industry.

So energy and utilities sector, they tackle a lot of not just inventory management, but resource management type problems. So they tend to work with large scale networks, problems that are heavily constrained.

These are typically good indicating factors that the problem itself becomes quite complicated to resolve relatively quickly.

So the good point here is that we are not limited by the problem size itself very quickly. We can find ourselves in a regime where the problem is too complex to be solved with existing exact methods.

So what we employ are what we call heuristic or approximative methods. They are widely used, have been widely used for the past decades in various fields, but they do not provide, or they do not necessarily provide perfect solutions, hence the name. They provide good quality solution, but there is typically room for improvement.

And an additional message that lends well for quantum computing based implementation is that they typically provide a limited number of good solutions.

With quantum computer you can expand this sort of portfolio of solution candidates. Therefore you can improve the efficiency by not necessarily improving the solution quality of your particular problem, but you can improve the overall workflow.

You can think of it as you’re tasked at allocating your resources for a specific delivery, but behind that, that delivery will be processed by a specific site.

There might need to be another delivery with a different type of van, boat with plane. So for your specific problem of just allocating your parsers, your packages to specific delivery sites, you might not know the whole picture.

So for these unknowns, getting access to many solution scenarios can be beneficial in improving the overall efficiency of the procedure.

[00:53:57] Speaker A: Thank you. Christian. We are down to about our last five minutes and I know we have two quick questions, so let’s try to squeeze them in. John, your quick questions.

[00:54:08] Speaker F: Yeah, first question is, is when you, when you build one of these quantum computers, are they specific to a single algorithm or a type of algorithm, or are they general purpose?

[00:54:20] Speaker C: They are general purpose machines.

So they are not yet what we refer to as universal quantum computers, but they are general purpose capable of executing a large quantity of algorithm of quantum algorithms.

[00:54:37] Speaker F: And the other question I had is if you have 100 qubits and you look at like the size of the asymmetric encryption keys, has your company started switching over to the NIST approved kind of quantum safe encryption algorithms?

[00:54:56] Speaker C: So I will be very, very brief on this.

So quantum safe encryptions do not necessarily imply using a quantum computer to protect yourself.

So it is part of the field, what we call post quantum cryptography, which relies on cryptographic methods that can be executed on classical machines that are proven to be safe against potential cyber attacks of even future robust quantum computers.

So here you do not necessarily need quantum computer to protect against quantum cyber attacks.

The whole normalization process is still ongoing.

So there’s a lot of attention for the NIST analysis. I think that should end later this year or early next year on finally identifying which algorithms are have their label of being quantum safe. But it’s still an ongoing process of investigation. What encryption methods are truly considered quantum safer?

[00:56:09] Speaker A: Thank you, Christian.

Let’s move on to Joanne. Quick question from you.

[00:56:15] Speaker D: Yeah, quick question.

If you were going to give advice to an enterprise that’s looking at this eagerly, anticipating the future and thinking very future forward, how would you create the bridge between.

Let’s assume that they’re not at a huge maturity level of AI, but mid level of AI. How would you bridge between AI and quantum?

[00:56:44] Speaker A: What’s the AI use case, I think

[00:56:46] Speaker D: is a, well, the AI use a large supply chain or materials or drug discovery or you know, any of the more complex engineering challenges that companies have.

Cold chain might be a good example.

[00:57:02] Speaker B: Yeah, I think the gap there, it’s, it’s not one of the other.

There is significant overlap between AI and quantum.

Yeah, you know, that’s sort of the best way to describe it is just that complementary. You’ll have AI jobs that can come up and be run on the cpu, GPU and then some of the data aspects, especially when you’re looking at large language models using, you know, machine learning techniques to be run on a quantum computer, then bringing them back into AI. So I think they work very much together.

[00:57:38] Speaker D: I’m, I’m glad you said that because I see it the same way. I don’t think it’s A one or the other. And I, I think that as AI progresses, quantum will be just added on and evolve with the two together.

[00:57:54] Speaker B: Yeah, a common misconception, I know we’re kind of bumping up on time is quantum computing is not going to replace cpu, gpu, FPGA or anything else for that matter. It’s just a continuation of the high performance computing stack. It’ll be suitable for some jobs, but not all jobs. So it’ll just be another, you know, silver bullet in the arsenal, if you will, of enterprises to drive results and earnings.

[00:58:22] Speaker A: Thank you, Mike. I mean, that’s a great closing statement. Just great way to think about this. I learned a lot today from you guys. Thank you. Michael and Christian from joining us and for Joanne, John, Martin and Joe’s questions. Michael, maybe just give us a very brief overview. What does Pascal do in the quantum space?

[00:58:40] Speaker B: Yeah, so Pascal, briefly. We’re based in Paris, we have 300 worldwide employees. And what we do is we harness the power of neutral atoms to build scalable high performance Quantum processors, or QPUs.

Our goal is to bring quantum computing out of the laboratory, away from the spooky science, into real world applications that can accelerate innovation across the different verticals like finance, materials science, healthcare, any one number of verticals there.

This, this approach has led us into a number. We have about 40 customers right now around the globe.

Very cost efficient, I may add. We’re not talking about. You always hear about chilling qubits to, you know, you know, minus 270 degrees Kelvin. The neutral atoms that Pascal cells are QPU and our cloud solutions run at room temperature. They drop into the traditional HPC footprint. So that’s something that’s been very appealing to companies.

[00:59:49] Speaker A: Wow. We’re getting a lot of applause here from Keith, David, Heather, AJ, for just a good session.

I learned a lot. You know, as you probably could tell, I came in a little bit skeptical, but quite frankly, if I’m hearing I could buy a machine for $20 million. If you work in an enterprise with an R and D budget north of $100 million, this is probably something you need to be looking at. If you’re working in industries with large scale computing problems that are already investing in high performance computing, you probably need to be looking at this.

You know, when Heather is on here and Joanna and I are talking about this, we always talk about hiring diverse candidates into our IT organizations. Maybe it’s time we start going back and hiring some, some physics students coming out and saying, look, you know, I want you to learn, learn and be our internal expert on quantum computing. So a lot that can be digested from here. I really like this, these use cases that we shared that if you’re working in any of these industries and you’re working in this space, do let me know.

I’m in sort of incentive now to do my own homework and see if I want to cover more around quantum computing in my writing and my research. So thank you again Michael Christian for joining us. Thank you Martin, Joe, Joanne and John for our questions. We’ll be back next week for the coffee with Digital Trailblazers. We’ll talk about Data Privacy week and we’ll be talking about how you should take control of your data, lining up some new experts for that one. On the 7th we’ll be talking about establishing product management for non in non tech industries. On the 14th we’ll be talking about embracing new roles and skills to support your AI error transformation. And then on the 21st we’ll be getting into intelligent operations. Folks, thank you for joining this week. I look forward to all of you being here next week. You can watch the recording on LinkedIn. I’ll have it up on my website. And for those of you who haven’t found this yet, Please just visit drive.starcio.com community I launched the Star CIO Digital Trailblazer community last year. We launched three new advisory Connect programs just over the last few weeks. And if you’d like to know more about it and have questions about it, do reach out to me on LinkedIn. Everybody.