Napoleon Cornejo, quantum computer expert, engineer, founder and director of tech firm Geokapti, shares the reality, benefits and challenges of quantum computing’s exciting prospects. We also apply the idea of QC to AI, discovering how the two can work in tandem for even greater power, and of course, how the customer experience will be affected in the future as a result of these advancements.
Quantum computing is an exciting field that has been in development for several decades, with the potential to revolutionise the way we solve complex problems. The idea of quantum computers was first introduced by Nobel Prize-winning physicist Richard Feynman in the 1980s, but it’s only recently that we’ve seen significant progress in bringing this technology to life.
Quantum computers use qubits, which can store a range of possibilities beyond the binary 1s and 0s of classical computing. This makes them ideal for solving complex problems that require a large number of calculations. Although still experimental, quantum computing is expected to have practical applications in industries such as finance, logistics, and scientific research, as well as for solving optimisation and machine learning problems. While it may not become a household item, businesses may soon be able to access quantum computers through cloud services to solve specific problems pertaining to customer experience.
The combination of quantum computing and AI could revolutionise the way we approach problem-solving. According to Napo, a quantum computer’s ability to compute multiple states at the same time could significantly enhance AI’s capabilities. With traditional computers, you have to go through each state one at a time and compute variables. However, quantum computers can perform these computations simultaneously, exponentially speeding up the process.
So how does this relate to the customer experience? Well by utilising this technology, businesses could gain valuable insights into their customers’ behaviours and preferences in real-time. This data could then be used to create personalised experiences that cater to individual customers’ needs. Imagine a world where chatbots could communicate with customers using human-like language and understand their intentions on a deeper level. This is just one of the many possibilities that could be achieved with quantum computing and AI.
This article summarises podcast episode 89 “CX in the World of Quantum Computing” recorded by CX Insider. For more information, listen to the episode, or contact Napo on his LinkedIn profile.
Written by Marcell Debreceni
Full Episode Transcript
Napo: There’s a famous quote that says, I’m not scared about AI. I’m scared of…
Greg: Space travel?
Napo: No, no. Creme brulee.
Greg: Thank you for coming all the way from the Netherlands.
Napo: A lot of things I’ve regretted. First of all, to meet you [crickets chirping]. Some studies say the human brain is actually shrinking.
Greg: Yeah, I’m adding to the study. Definitely backing up the theory. Yeah.
Marcell: Hey, everyone, welcome back to the CX Insider Podcast. Today we talk to Napo Cornejo, founder and director of Geokapti, a tech company that innovates software solutions using satellites. We’ll learn about current and future technological advancements that will change customer experience like quantum computing. And I enjoy the conversation. And if you do, why not subscribe to our YouTube channel for CX Insider’s Best Content? By the way, this podcast is brought to you by ACF Technologies, Global Leaders in Customer Experience Management Solutions. All right. Thank you very much for coming on the podcast Napo. Pleasure to meet you and have you on here. To get started, would you like to tell us a bit about yourself, who you are and what you do?
Napo: Yes. Well, first of all, thank you for having me here. Pleasure to meet you. My name is Napoleon Cornejo. Rather unusual name, but I come from El Salvador and now living in the Netherlands. I’m originally a software engineer, now the head of a company called Geokapti in the Netherlands, where I’ve been living in the past 15 years or so, we work doing mostly scientific software research, software engineering, software, dealing with AI, quantum computing, aerospace and other high tech industries. Yeah, and that has brought me here today. Cool.
Marcell: Okay. So that’s obviously quite a lot of different areas that you’re involved with there. What are some of the most exciting projects that you’re working on right now or maybe something you could share?
Napo: Yeah. So we are working currently, for example, in a project that processes large volumes of satellite data to detect soil moisture in Europe. So we take data from radars that are in satellites. These orbit the earth periodically and we take the signals from this radar to say something about the moisture of the soil, in this case in Europe, because the soil absorbs these signals in different way when it’s dry and when it’s wet. And this information helps agricultural policies, conservation, etcetera. And the other project that maybe I can mention is one that has to do with quantum computing. Here we are writing the software to control the smallest components of the quantum computer called qubits. So we help a research institute in the Netherlands to achieve that sort of control. And then besides that, we have other projects related to artificial intelligence machine learning, some of which we are working with ACF.
Marcell: I used to think quantum computers were just this kind of sci-fi dream, you know, something in the future. And computing obviously has come such a long way from its inception. Could you share what a quantum computer actually is, what that means, what it’s capable of? Yeah.
Napo: So indeed, there’s a lot of, a lot going on around quantum computers today, but the idea actually comes back to the end of the 70 seconds, early 80 seconds. We didn’t have the technology there, but that’s when the theory really got started. A prominent physicist who talked about this idea was Richard Feynman. He won the Nobel Prize in physics. He posts some of the first principles of quantum computing, how you could make computations with quantum mechanics. And now what we’re trying to do is actually bring that into reality. So these quantum computers are composed at the very end of what are called qubits, which are the analogous of classical bits, which can store ones and zeros. But these qubits can store a myriad of other possibilities. And it’s in that myriad of other possibilities that you can get the large amounts of computation. Of course, there are challenges making stable qubits and that don’t fade away with noise is extremely difficult and probably that’s the biggest challenge at the moment. That’s where the software that we’re writing right now comes in, in trying to control these qubits, to keep them stable as long as possible so they can actually do something useful.
Marcell: What are some of the possibilities that it could bring to life in terms of achieving different computations?
Napo: Yes. I think I should start by saying that there’s a bit of misconceptions in terms of quantum computing. Some people think that this will actually replace the computers we use today. That’s not true. It will be a complement to classical computers, so it will not substitute them, but they will be dedicated to solving very particular problems, especially problems where you need to try hundreds or millions of possibilities quickly. This is where quantum computing really excels because as I said before, a lot of possibilities can be encoded in a single qubit at the same time. So the values that it can store at the same time are infinite in theory. And therefore you can search and compute with much more values in less time. So for example, optimisation problems where you need to search a lot of possibilities. What’s more optimal, what’s not based on hundreds or even thousands of variables, then quantum computing can be here to actually compute all these possibilities rather quickly, which a classical computer could take years to compute. Yeah. Um. Machine learning where you need to explore data, create models, taking into account thousands of variables or hundreds of thousands of variables or millions. Quantum computing can help there as well. These can further be applied in industries like finance, logistics, mathematical modelling, scientific research. Et cetera. Et cetera.
Marcell: You say it’s more of a complement to everyday computing, do you think it could be quite a practical thing in the near future that although it won’t replace normal computers, it will be quite common or widespread?
Napo: Not at the moment. Right now it’s quite experimental and very new. And again, there are a lot of challenges to make it practical. At the moment, the only quantum computers that exist are in labs with extremely well-controlled temperatures that almost reach absolute zero. These need to be controlled and manipulated with lasers or microwaves. So it’s all these big refrigerators just to store a tiny little chip. But it takes a big refrigerator control with liquid nitrogen and so, so forth to make it work. So at the moment it’s just experimental. They live in laboratoriums. It’s not practical yet, but as all of technology, right, it begins this way. Sure. Just like the current computers began this way. Yeah, exactly. As very big building-size computers, which then became smaller and smaller until you have a cell phone in the palm of your hand. So technology starts this way. Quantum computing is just at the beginning, so it’s rather expected that it is like this. As years go on and time goes on, I think there’ll be good progress made. I cannot make predictions of when it will actually become. That was going to be my household. A household item, but maybe practical uses are not that far away. 5, 10 years perhaps still carried out in labs, but solving already important problems. Right.
Marcell: So maybe not a household item, but maybe businesses could start using it for improving their customer experience.
Napo: So for example, cloud services, they could have their quantum computers in their own installations. Right. And just hire it over the cloud to solve specific problems. That’s how I think it will start. It won’t be a quantum computer in your house, perhaps.
Marcell: Yeah, maybe one day. But that’s really fascinating. Surely there’s also quite a few risks that come with that. Post-quantum encryption. Yeah. Could you explore that idea and what it means and are there any potential risks?
Napo: Yes. So as with every technology, of course you could use it for either good purposes or bad purposes. I mean, you could say that about fire, even. So, quantum computing is not the exception. One of the areas that’s being explored is indeed encryption, because current methods of encryption are based on the fact that finding the solution to the encryption problem takes millions of years for a classical computer. There are just so many possibilities.
Marcell: Like with passwords.
Napo: Yes, with passwords. But this one even more difficult because they are mathematically complicated and using numbers that are hundreds of thousands of digits long multiplied with several operations and cycles. So you need to find the right combination of these numbers to actually being able to decrypt the key. It’s extremely difficult and takes a classical computer millions of years to solve. But as I said before, with quantum computers, these qubits are able to encode a myriad of possibilities at the same time in the same qubit. So it’s calculating all these possibilities at the same time, or at least in theory is able to do that, which means that guessing all these combinations of numbers and operations to unencrypt your key basically in a brute force way could be done realistically within a time that is actually threatening to your bank.
Marcell: Security concerns.
Napo: Yeah, exactly. That’s why part of the research that’s being done in quantum computing is what are the ways encryption can be enhanced to avoid this. So that’s what we call post-quantum encryption. So encryption schemes that will be able to withstand the computations done by a quantum computer.
Greg: That’s a very real risk, isn’t it, for us, and one that has to be solved because like you mentioned there, banking. People’s money sits on top of technology that utilises what we what you’re referring to there as like sort of the current encryption, whereas quantum computing would be able to solve that within a short, shorter time frame, exponentially quicker time frame, let’s say.
Napo: Certainly, or communications, your WhatsApp, your signal messenger, these are all based on the principles of encryption. These are broken. Then your communication is exposed.
Greg: What do you think are the first uses of quantum computing that we will see?
Napo: One of the first applications that’s already going on is quantum communication, where you can really entangle two particles and actually make them communicate qubits, entangled qubits. This communication is ultra-secure because it’s based on the principles of quantum mechanics. That means that an eavesdropper will disturb the state, making it impossible for the communication to go on further. So they are already testing the first quantum network to make this type of communication and they are already deploying a first prototype of this. And the other application that’s already going on is modelling molecules. For example, when you want to create a specific type of medicine at the very molecular level, these are based on shape how proteins fold and how they match with each other and so forth. It is extremely difficult. It is a rather difficult problem to solve with a classical computer. Quantum computers, which already work with quantum principles that are the ones that play on the molecular level, they can solve these problems much more easily. Okay, that makes sense. So and there’s already papers and literature on how this could be solved, at least in theory. Once these qubits are there and find out new molecules for therapies, this will come into play for finance, for example, calculating risks on real time, using worldwide data and at every moment calculating your risk and knowing your risk position. Tremely valuable logistics problems on the world wide level shipping, commerce. These are all applications that are already being thought of space travel. Space communications. Communications. Okay. Travel maybe not yet. But space communications. I mean, the Chinese already have a satellite that uses quantum communication.
Greg: Wow, a lot of applications coming in. Yeah. Watch this space.
Marcell: There is obviously quite a lot to explore with quantum computing, having given an overview of its applications and challenges. Now let’s link back to a more familiar development. Artificial intelligence is also a huge piece of tech that’s already changing the way we work. Personal assistants like Siri and Alexa have been around for a while, but services like ChatGPT and Dall-E are pushing the boundaries further as both AI and quantum computing grow. How might the two intertwine? Can they be combined for even greater power and potential?
Napo: So there are several ways in which this interaction can happen. So how can quantum computing aid? AI is indeed because of the possibilities of computing with large amounts of data and the qubit being able to encode such a myriad amount of states and compute with them at the same time. So whereas a classical computer has to go through each state one at a time and compute the variables, optimise neural nets, just neural nets or whatever, the qubits or quantum computer could be doing these possibilities. At the same time, several of them.
Greg: At the moment, I guess we’re impressed with what we can see from Chatgpt and technologies that use AI like this. And I guess what you’re sort of saying there is that if you combine quantum computing with AI technology, we could see something exponentially even more powerful than what we already are very impressed by.
Napo: In principle, yes. Take ChatGPT for example. ChatGPT – and you’ve seen the results – it produces text that is very human. Like I mean, it’s something like a human being could write, but at the very end, ChatGPT works on calculating probabilities based on a neural network that’s been properly tuned with a lot of text and examples and human-generated content. And after being trained, this massive neural network, which is about 175 billion parameters, it produces what you can see. Take into account the amount of resources to make all this computation, to do the training data, the time it takes. If you can already see that combination with other technologies, it’s just going to get more and more impressive. ChatGPT is just a neural network that’s guessing the best next phrase or the best next word. Now, that brings a bit of a philosophical question, right? Because, okay, if that’s the only thing that it’s doing but produces human-like text, is that really like the way the human brain actually works then? Do we actually work a bit that way, that we think about something and then try to guess what’s the next likely follow-up to what we’re saying? So it brings up all sorts of interesting questions beyond computer science.
Marcell: Yeah, because it tells you doesn’t it? Like I’m just an AI. I don’t really I don’t have opinions, but these are kind of the facts or this is what I can construe from the data. And then you just have to kind of run with it. And I suppose that’s why that human layer of interpretation will always be there as that layer on top. I suppose with quantum computing coming into the conversation as well, I feel like it’s quite hard to actually imagine what the uses will be and how we will be using these technologies in the future. Ten, 20 years ago you wouldn’t have guessed what’s going on now.
Marcell: It is great discussing the future of technology and how it will reshape the business world, The question of ethics looms in the background like some shapeless shadow, because there are surely many benefits. But it’s also easy to grow anxious about the inevitable drawbacks of rapidly implementing this kind of tech. So will humanity adapt in time, or will we get left behind?
Napo: I think that human beings will adapt. I can be a very powerful tool that could make our work more efficient. If we learn how to use it properly, it can automate a lot of things to give us more time to actually do. The creative stuff could substitute, I guess, some jobs. Others it will make it more efficient and humans will have to adapt to that. You don’t see people working in telegraphs or in switchboards. Your generation had to adapt. Yeah, there are new jobs where technology helps you become more productive. You saw this letter put out by several companies or CEOs and scientists on having a moratorium on AI because now with ChatGPT, people are starting to get scared. It could be damaging for humanity, but it’s not realistic. Someone somewhere will continue to develop AI, a government, an organisation, a single individual in a research institute. If you make a moratorium on AI, what you’re basically saying is, okay, we’ll pause so that you can get an advantage, right? Yeah.
Greg: What do you feel about artificial intelligence making decisions in more critical situations like in healthcare?
Napo: For example, there’s a lot of research on AI that interprets x-ray and actually in some cases it comes to make a better diagnosis than actual specialists. But it also does make mistakes. And the problem is, can you hold the machine responsible for the life of a human being? Can you take it to court? Where is the liability? So AI is a tool to help us become better. And in this particular case, like health care, it should always be vetted by a human specialist, a doctor or a council with the AI as an input, but not as the final decision maker.
Greg: Yeah, maybe that’s the morally best place for it to stay for now.
Napo: Right, indeed. But take, for example, cases where there’s no chance to do a review, automatic driving, for example, in a car, whether you’re in a situation where the car has to go somewhere and there’s a person in front, but if you move it somewhere else, there’s two persons on the other side. So yeah, yeah, it’s that. What should you choose? Right?
Marcell: Stop the car, I suppose.
Napo: Yeah. If that’s at all possible. Or in the highway or so forth. But even in a normal situation, if for some reason there’s a mistake, there’s a flop in the computer or whatever, and you run over someone, can you hold the machine responsible? Who’s the responsible party there? Is it the maker of the car? Is it the software developer of the AI? Who is or is it the driver or is it the driver? So this is a bit more tricky. Yeah, if there’s a driver, but if the car is fully autonomous, it’s even more difficult.
Greg: A question I’m always asking myself is why don’t we have technology where you can speak to me in Spanish and I can hear it in English and I can speak to you in English and you hear it in Spanish in real-time. How far are you away from that technology?
Napo: Actually, that’s not far away. I’ve seen some prototypes of that. YouTube has automatic translation and there’s already software that can synthesise voices out of text, putting all those parts together. There are prototypes. I don’t know of any commercial company that maybe there is. I just haven’t noticed. But I know there are prototypes of that. We are so used to it that we don’t see a miracle in it anymore. It’s an everyday thing, but it actually took a lot of effort and a lot of research to get this translation software to get working. And I’m sure the same thing will happen to this voice. Synthesisers plus translation in real-time and so forth.
Marcell: So in a way, do you think that’s maybe like encouraging or letting people learn languages less? Like, you know, because if you’re trying to speak to someone, you have that technology that can just give you that instant access to that information. Do you think it takes away, like the whole education and learning aspect of learning a whole language? Because obviously learning a language and things like that are so beneficial to the brain, right? From a wider perspective as well. Do you think like integrating all these automated technologies into society will have negative impacts?
Napo: There are studies being done that says that the human brain, the size of the human brain, has actually shrunk. Human beings from 10,000, 20,000 years ago who had to memorise where was the tree for apples? Where was the way to hunt? I mean, they knew their environment very well. All of these sorts of skills that they had to learn and be quick about them, which we don’t need anymore. Yeah, we don’t need anymore. So when you say about languages, indeed, I do think that because a lot of people learn languages out of necessity, they actually need it to speak another language. These technologies will take out that need out of the equation so there’s no real push or need to learn it. It’s only if you want it, but not really because you need it. And the same happens with other technologies. Take maps, for example. Yeah. Before Google maps to take the paper and actually know how to read it right. And look for the actual street and even oriented the right way to find your way around cooking, starting a fire, whatever it all required skills and brainpower. And that’s why when you say is it having an impact on the brain, some studies says that yes, the human brain is actually shrinking.
Marcell: Wow. That’s a bombshell to drop there.
Valentina: [in the audience] I have a question about the study on the human brain shrinking. In what timeframe was it done?
Napo: It’s tens of thousands of years. No, it’s not hundreds of years. No, no, it’s tens of thousands of years. Indeed.
Simon: [in the audience] I think it’s shrunk over here more than anywhere else.
Greg: Yeah. I’m adding to the study. Definitely backing up the theory.
Marcell: Yeah, I think there is that danger, though, isn’t there? Like if we don’t use our brains because everything is done for us, then what’s going to happen?
Napo: Precisely. Indeed. And I think that that’s one of the drivers of this phenomenon that we don’t need to use all the skills that previous humans had to use.
Greg: So all the sci-fi films are wrong. These aliens don’t have massive heads. They have tiny heads!
Napo: Yeah, maybe. Maybe.
Greg: And we’re going to be walking around the future in these tiny heads.
Napo: Indeed. No, that’s right. That’s right. That’s funny.
Marcell: And let’s bring the conversation home. How will quantum computing and artificial intelligence alter the customer experience? How will these things specifically impact the business world? And why should companies even care? Napo shares his thoughts on the new age of customer journeys.
Napo: Yeah, you can easily imagine how ChatGPT will impact the chatbot technology where you can train it with your own material, with your own marketing material, your own documents, your own history, and then produce a chatbot that actually has the knowledge to respond properly to questions made by customers. Imagine how many customers you can now answer questions to which before you needed a customer agent to be responding each of these questions, right? Yep. Yep. So that’s already one big thing. The other is optimising the customer journey, for example, estimating the proper attention times or who should handle his case, assigning the proper resources, having models that predict the demand of customer service as well. So all of these goes straight into the applications of AI could have a big impact in this industry. Everything I said about quantum computing that will actually make it even more powerful than you can imagine already how this will change.
Greg: And it’s good to talk about the fact that you work with ACF Technologies today with us on our own artificial intelligence technology and how we’re planning to help our customers and future customers transform customer experience using that technology. Do you see AI playing a role not just in the customer-facing side but also in the back office side? So you know, where operationally organisations are managing the availability of resource, you know, back office decisions. Is it there as well?
Napo: Certainly, certainly. And that’s one of the things we work on optimising processes. I can build models based on previous history and previous results to find out what are the proper variables to make a decision, who should take a hold of a case, what resources you should assign to a certain amount of cases, predict the resources you will need for that day. Optimise the scheduling of doctors, nurses. That all happens in the back end. It doesn’t happen on the face of the customer. It really makes his experience much more enjoyable, more efficient, with less burden. It really makes an impact.
Greg: If we think about what great customer experience means, especially in the future, and I think we’re working towards it. Customer engagement with any organisation. Customers want to engage on the channel at the time in the means that they feel is best for them, you know, and it’s sometimes it’s in person, sometimes it’s via chat, could be via video. I feel like the future is only driving in one direction, which is that customers are going to be more demanding on what they want and they’re going to get exactly what they want. And maybe technology like AI is going to be the only way for an organisation to deliver to that expectation.
Napo: I think it’s indispensable. There won’t be a way around it. Companies that really want to provide an enhanced customer experience should embrace this and become familiar with it and learn to use them.
Greg: Okay. These peripheral, very advanced technologies, quite commonly, it’s engineers talking to engineers, taking that conversation and converting it into business language and business benefits for customer experience. How does an organisation go about that?
Napo: $1 million question for engineers. It’s really difficult to communicate with non-engineers, with commercial people, with marketing people, because we’re used to speaking in the language of equations, software, algorithms, functions, files. That’s the language we speak with. It’s a very deterministic, concrete, static language, commercial people, marketing people. They deal with other human beings. If an organisation really wants to take a leap in these new technologies, which they should, they should embrace them because the world is moving in that direction. They should really strive to get people that can serve as a bridge between these two worlds. Having people that can bridge these two worlds is an extremely, extremely valuable asset.
Marcell: Thanks for listening. We hope you enjoyed the podcast and if you did, why not subscribe to our YouTube channel for access to full-length videos and YouTube shorts. You can also like share and comment on the episode to keep the conversation going and be sure to check out cxinsider.com for more content. Now, if you wish to join our growing community of thought leaders, head over to LinkedIn and follow us at Insider podcast to stay updated. Thanks again. I’ve been Marcell and I will see you in two weeks, but for now, enjoy our Rapid Fire questions. By the way, this podcast has been brought to you by ACF Technologies, Global Leaders in Customer Experience Management Solutions. So my first question is what is your favourite thing about your job?
Napo: Working with scientists. Because I learn a lot from them and the science.
Marcell: Yeah, I’m sure that’s really fascinating, especially for someone who understands what they’re saying as well.
Napo: Well, to a certain extent, yeah. I still have my questions, but it’s fun to engage with them and see what they’re working on and they’re really at the cutting edge of knowledge. Yeah.
Marcell: Is there anything that you’ve regretted buying the most?
Napo: A lot of things I’ve regretted buying. Probably a bed. The cheapest that I could find, but it was really, really bad. So I really regretted that I should have invested more in that. But yeah.
Marcell: Invest in your sleep.
Napo: Yeah, definitely. Exactly.
Marcell: If you could transform into any animal, what would it be and why?
Napo: It definitely would have to be a bird. I like the freedom of flying. Maybe an eagle, perhaps.
Marcell: And what’s your favourite dessert?
Napo: Creme brulée. I love it.
Marcell: Nice straight answer. We can tell it is true. Yes.
Napo: Off the top of my mind. Yeah.
Marcell: And if you could interview anyone, dead or alive, who would that be?
Napo: Isaac Newton. Probably just being able to tell him what has happened since his theory of gravity. I think it would blow his mind.
Marcell: Do you think he could comprehend anything?
Napo: I think he could. The way the story of how he came up with his theory of gravity is really impressive. So I think not only understand it, but he would be just fascinated.
Greg: Favourite holiday destination?
Napo: That I still haven’t gone to is Nepal. I want to go to Tibet. I want to see the Himalayas with my own eyes.
Marcell: Would you climb any of them?
Napo: No. I’d probably die if I tried. I don’t have the training.
Greg: Skip that.
Napo: I’ll skip that just as long as I can see it. Yeah. Appreciate it from afar. I mean, they say even getting to Camp Zero is already a massive effort. Yeah. So looking from afar.
Greg: Great. Any other questions from you guys in the audience?
Simon: What scares you the most about AI?
Napo: What scares me the most about AI? There’s a famous quote that says, I’m not scared about AI. I’m scared of people. People. They actually make decisions that impact on the way we use technology, right? They are the final decision-makers on how this technology is used or how is it overseen or what are the regulations. And if that’s not done properly, then that can have an actual impact not only in the company but also in the country. I’m more scared of people than of AI.