Dario Gil 

Alright, good afternoon and welcome to exploring the quantum computing future that we're creating right now. And it is a real pleasure to be here with with all of you to discuss and explore some of the implications of these amazing technology. Let me set it a little bit on context on the institution I represent, and that I lead for IBM IBM Research. And this year, it's actually our 75th anniversary that we've had our research division. And a president, we have over 3000, scientists and engineers distributed all over the world. And our history, we've maintained always this commitment to do very fundamental investments in core science of the type of work that have produced Nobel Prize winning achievements, but also to explore the implications that these advancements would have to the world of information and the world of computation. Just to give you a little bit of a flavor, where some of the work that I'll be sharing with you today gets carried out. This is the laboratory where I work. This is the TJ Watson research lab is about 45 minutes north of New York City and in the building, we have about 1700s Scientists and engineers that work here, and I'll show you also, you know, results along the way. This is our laboratory in Almaden in California, a little bit south of San Jose, who's on in on top of this beautiful state park. I'll show you just one more laboratory. This is a laboratory in Zurich and crucially con, where a lot of very fundamental physical sciences work also gets carried out for IBM Research. And I share this because, you know, in the context of what we're going to be discussing, what we do is we attract extraordinary mathematicians and physicists and chemists and especially in engineering and in quantum as an example, to explore what is next. And this is the most exciting time in the world of computing in the last 50 or 60 years. And simply put, what is next is we have this intersection of technologies are coming together and have these very simple equation to define what is happening now and what is next computing which is Going to be this intersection of bits, plus neurons plus cubits coming together. And that if we were to sort of very briefly summarize the last 50 or 60 years of information and computation, we could say something along the following lines that we build the information technology technology edifice, on top of the ideas of cloud Shannon, that gave us what is now known as classical information theory. And Shannon is the one that taught us to think about this world of information abstractly. He told us, he introduced to us the term of the bit, the binary digit. And he said, Well, we could the couple in some ways, this idea of zeros and ones from their physical implementation, and is that separation that allowed us to go and see things like a punch card on DNA, and come to appreciate that something physically different as those two things had something in common, that both can be understood as carriers of information and we This and as a companion with Moore's Law, we have built modern computing. With digitize the world. We've made bits ubiquitous, and essentially almost free.

 
Dario Gil 

And the situation that we now face is the consequence of the intellectual quest of a series of scientists that beginning in the 1960s took that very idea that Klaus Shannon put forth the separation of physics and information and started to poke at it. They start to question whether it was really true that physics and information were separate. And I'm showing here an example of Charlie Bennett, for example, on the right hand side, who is one of the fathers of quantum information theory. And what this physicist started on in the early 60s and 70s. They would ask questions like this, they would say, is there a fundamental limit to how energy efficient computing can be? They would ask questions weather information process Could be thermodynamically reversible, only the kinds of questions that are physicists would ask right about those kinds of things. And poking down that thread began to understand that there is another way to understand information. And this is a photograph of a notebook from Charlie Bennett, that has the words, look it up there quantum information theory, we think is the first time that those words were written down. And that was done in 1970. In our lap, so that's how long we've been in this journey in IBM Research exploring these ideas. And what we came to appreciate and realize based on that theoretical work is that the building block of the world of information is not the bit the zero on one, but rather something known as the quantum bit, the qubit for short. And that at the heart of this idea of the qubit, we're going to bring three ideas of physics that is going to enrich the possibilities of what we under That information to be, and the three ideas, ideas of superposition, interference, and entanglement. So let me just briefly describe what they are and why they matter. So superposition, it's actually very easy to understand we have a classical analog to it. So in the classical world, we have zeros and ones, let's say we had a coin, heads or tails. And when the coin is spinning, if I ask you, is it heads or tails? What would you say? You would say, yeah, it's a combination of heads and tails. When I perform the final measurement, I either get one position or the other. Okay, so So far, so good. So now we're going to introduce the second idea. And by the way, so that idea in the quantum world for us are the zeros and ones is that instead of having a zero or a one, only, I could have a superposition, combinations of zero and one. The second idea is actually really profound and sophisticated. So I won't spend a lot of time I'll give you an intuition. Fred. And there's a idea of entanglement and entangled cubits in this case have this property that it is impossible to describe the combined state independently of one another. So let's go back to our coins. So in this case, we're going to have two coins. And in classical world, if I was spinning the two coins, and I perform a simultaneous measurement, whether one coin was heads or tails is independent of whether the other one is heads or tails. The probabilities are independent. In quantum world, if these coins were entangled, when I perform that measurement, now I can have situations when one is heads, the other one is always heads, and one is tails. The other one is always tails. That's one of those things that when you actually reflect on it is one of these Earth is not flat kind of moments, right? That is a really profound thing because what it tells you is you can have non local action in the universe. So this property which is weird in its nature, but it's something that reflects the nature of physical reality, there is a relationship between this special property and the world of information. And it's the following. If I had 100 perfect cubits that were entangled, that had this property with one another. If I try to represent all the possible states of that entanglement of those hundred cubits, using bits, using zeros and ones, I would need to devote every atom of planet Earth to store zeros and ones. By the time I have a quantum computer with 280, perfect cubits, I would need every atom of the known universe to store zeros and ones. So that's a very interesting property to go on harness and see what we can do.

 
Dario Gil 

The third principle of physics that we bring into information is the idea of interference. And this is something again, we're all accustomed to seeing in waves in the ocean. And the idea is you can have waves that can interfere with one another and constructively add up to form, you know, the peak of the wave. And sometimes you can cancel the forms the valleys of the waves. So those three ideas. Okay, so how does an algorithm work? to just give you an intuition for how different it is than a classical computer, in which you have switches of zeros and ones, and you implement the basic logical operations of the ands and ORS and the knots? So how does it work? In a quantum computer, the first thing we do is we put the quantum computer in a superposition. So let's say we have a very simple quantum computer we just do cubits. The number of states that is available in my quantum computer is due to the power n where is a number of qubits. So since I have two cubits to to the power of two is four, so I have a 00011011 state. In this case, we're going to represent them in a sphere. So imagine that if I wanted to put a one, I would put a.in, the north pole of the sphere. If I wanted to have a zero, I could put it in the South Pole. And if I want to have a combination of 01, I would put a.on the equator. In this case, I have four dots in my sphere each.is represented by an amplitude and a face to face is where is it in this sphere in its rotation? Okay, so first thing, that's all I do. Second thing I do is I need to put data into the computer. What it means in practice is that is when we send that information into the computer, we change the face where it is in the sphere of those dots. And here's where the special step comes in. The next step of writing an algorithm another I have my data into the computer is the process of taking those states that have an amplitude interface and interfere them with one another, like waves in the ocean to maximize the answers I want and cancel the answers I don't want. So here I have to the possibility of playing with negative probabilities, I can cancel things out. So I can explore a sea of possibilities to find the answers I want with this process of interference. Very, very different than a classical computer. Why does this matter? This seems so esoteric and theoretical. Well, the reason it matters is because we could divide the classes of problems that we can solve in the world into easy problems and hard problems. Easy problems are ones in which the number of variables we have to compute over are not exponential. And really, that's what classical computers, all the computers, we all have are good at solving.

 
Dario Gil 

We know there are other problems that are exponentially difficult to solve, because the number of variables are exponential in nature. Examples of problems like that look like that are factoring, which we use as the basis to encrypt a lot of information in the world. But very fundamentally to modeling nature, the process of doing chemistry and materials and understanding how nature behaves is also exponential in nature. And the thing to realize about quantum is that is the only technology we know that alters the equation of what is possible to solve versus what is impossible to solve. I'm not making the claim that all hard problems will be able to be solved with quantum computers. But I am making the claim that it is impossible to solve those hard problems with just classical computers. So that's really important and powerful. Can you build these machines? I made the claim that this is the most exciting time in computing in many decades. So useful to go back to this moment. So this is 1944. This is the first programmable computer that was built at Bletchley Park. And in some ways, when a similar moment where the first programmable quantum computers have been realized and built, and this doesn't happen that often, right, and it's very, very exciting that we find ourselves in this moment. And inside it, there's different ways to build these quantum computers. In our case, we use superconducting technology based on a device called a transman qubit. And what we have in in the end is we have a device that is about 100 nanometers. In dimension, it's just a sub junction and it's a sandwich of aluminum aluminum oxide. And basically we are able to create the equivalent of an artificial atom that creates a ground state and an excited state. What you're seeing in here is that that device, each one of those squares represents a qubit. And you see these wiggly lines that connect cubits to one another. And that is what allows us to these coupling resonators to build these superposition and entangled states, between different cubits. So in the you in the end, we're using silicon technology that we have benefited from decades of investments in semiconductor technology, and create a new device and a new capability to be able to create these quantum bits. Now, this realm of making these cubits and doing multi qubit demonstrations was the space of a few laboratories around the world that were capable of manipulating these very, very delicate information. But this change in 2016, where we were the first company in the world to build working programmable quantum computer in that case with five kids Bits and make it universally available through the IBM Cloud. This was the IBM q experience. And basically anybody could sit in front of a terminal, right up program clicks send. And you would be able to run your program in an actual quantum computer. This is what we did. So you write your quantum logic, your program, we send it that you send that normally as zeros and ones. And then we convert those zeros and ones to microwave pulses that operate about five gigahertz. These microwave pulses traveled down a cryostat, and at the very bottom, there's these cubits that operate at about 15 milli Kelvin. So it's one of the coldest places in the universe, right, the bottom of a quantum computer of this nature, we implement the superposition and entanglement and interference operations, then we amplify the signal that results and we converted back to zeros and ones and we give you the result.

 
Dario Gil 

So here's what happened since we launched that system. He sees the emergence of a new community of quantum developers and programmers and scientists. This is the users around the world using IBM quantum computers. And you're seeing here that progression over time since May 2016. Until today, and you're seeing both at the bottom. We gave people access to a simulator, an actual quantum hardware that you can run the executions. And what is very, very interesting is Look what has happened, relatively reason, the explosion of the use of the quantum hardware, the quantum systems themselves today, to be able to be at the forefront of the state of the art of doing quantum development, you got to have access to actual hardware. Right. It's not enough to have simulators. And what you're seeing here now is well over 130 billion executions have been run, and over 120,000 over 200,000 registered users are part of this community. One year ago, I was here doing the opening keynote with Ginni rometty, the CEO of IBM announcing Creation of the first fully integrated quantum system, IBM q system, one that you see here. And it's amazing the progress of what has happened over the course of of a year. At present, we have 15 deployed quantum computers serving now the community over 200,000 registered users, over 150 billion executions have been run in this fleet of quantum computers over 200 scientific publications enabled by these quantum computational hardware. And I'm pleased to announce today that now, we've crossed also the barrier over 100 commercial partners. And we'll have an opportunity to have a dialogue in a minute with two of them. And that's a major testament of the rate on progress. He's also the evolution of the 15 quantum computers and the different architectures that we enable in the IBM Cloud from five cubits to 16 to 20 cubits to a 53 qubit system that we announced also earlier this year that we have deployed And what is interesting about this, if you want to sort of understand the rate of progress from a systems perspective is that the number of cubes themselves, although important, it's a very incomplete measure of a power of a quantum computer. It's very easy to print more cubits. We have a list of graphic techniques in which we can do this. What's very, very hard is to make the cubits and the interaction between the cubits to be a very high quality. So a way to measure the power of a quantum computer is to a metric called quantum volume. And quantum volume is a metric that incorporates both the number of cubits and the error rate, among other factors. And what's really important is that the really difficult part is to lower the error rate of these multi qubit operations. So, one of the things that we're very committed is to have this discipline and the discipline of the roadmap. We've lived this in semiconductor technology right, where systematically we will doubled the performance of or semiconductor processes. And here's an example of multiple generations of 20 qubit processors. And what you're seeing here is the distribution of the gate error rate over the course of the last two years. We're getting them tighter and tighter and better, I'm better and this this idea of sustain engineering improvement of building these systems. So we have defined a new roadmap. Now remember, I mentioned at the beginning that we build the information technology edifice, on the ideas that cloud Shannon gave us. And I said that there was a companion idea which was Moore's law. Well, when Gordon Moore first wrote his famous paper, that we then associated with the law, he had four data points in 1965 recreated it here. And he had that observation that the number of transistors per unit area was doubling every couple of years, but four data points. So earlier this year at the American Physical Society annual meeting, we laid out results and this roadmap and we said that over the last few years, we've been doubling the quantum volume of the quantum systems in IBM every year, so instead of every two years every year, and we had done quantum volume of four, to eight to 16, so we have three data points. And today a second thing I'm also pleased to announce is that we just recently two weeks ago, double the quantum volume, get again, to now 32. So this is the fourth time we've done consecutively every year of doubling the quantum volume of quantum computer. So at this point, we're in a similar stage of the four data points. And we've affectionately called this gambetta as law for Jagan betta, who is our head of science and technology for quantum computing program and an IBM fellow who originally along with his team came up with this methodology and proposal for measuring the power of a quantum computer and we're committed to At least double quantum volume every year. And if we keep at that pace, we're going to see really spectacular results right as a result of that, what is happening in the years ahead.

 
Dario Gil 

So this is a little bit I want to do details, but the way we achieve that achieve that is that we have to be creative about the topology of how we make the cubits interact with one another, so that there's less crosstalk and interference when the cubits operate. We have to make the device physics better. So here's a measure of the coherence time, how much time you have a quantum bit before it becomes a bit. It used to be for superconducting technology that that number was one nanosecond, right 20 years ago, now resolves that we also some recently we've achieved 500 microseconds To give you an example of the increases around that. And here's an example of the how we measure the quantum volume 32. You're seeing that in the blue line of quality factors of close to 4 million now, and experimental results that we have where there's plenty of room to continue to make progress on this front. So we're very bullish on the opportunities around that. So let me frame it now of where we are in this journey of quantum. For many decades, as I've been sharing and arguing. We've had these quantum science investments, right from the very theoretical foundations to core physics ideas, to creation, a multi qubit systems, etc. We undoubtedly find ourselves in a different moment now. The Quantum science investments will continue, of course, but we're in this phase of getting the world quantum ready. a whole generation of developers are going to need to learn how we program these quantum computers. We're going to explore the applications and we're going to have to make these systems available to the whole world. And the goal is to achieve quantum advantage, which is the exploitation of quantum computing for scientific and commercial advantage. And that is absolutely going to happen this decade. So it's very important that we get ready, and that we push as aggressively as possible as us as possible towards quantum advantage. And that is why we launched the cube experience. And that is why we launched the IBM q network about a couple years ago. So to give you a sense of how much interest there is, this is this piece of news that we were sharing today. That now there is 100 members who are participated. And you can see a list here from industry partners and hubsan, startups and academia. And to just lay it out for a minute in a map. Just to see you the extent and the reiterate this. This we're talking about the academic institutions and universities and national labs around the world, who are part of the IBM q network are enthusiastically investing and exploring the opportunities of the technology. And here is in the commercial world while we're live. Looking at startups and industry and the broad representation across the world, so this fascinating industry interest when we started a year ago, you know, we were probably 35 to 40 partners today 100 and the rate is accelerating. So, this is tremendous momentum and excitement on quantum computing. So, one of the see if I can advance to the next slide, please.

 
Dario Gil 

So, I think we will go back one please.

 
Dario Gil 

So, you may be interested to explore what are some of the use cases that the commercial partners and the industry partners are exploring? So let me invite on stage now, Jamie Garcia, who is the senior manager for quantum applications, algorithms and theory, who works with many of these partners to explore some of the fundamental implications. Jamie, welcome on stage.

 
Jamie Garcia 

Thank you. Thank you.

 
Dario Gil 

So Jamie We have a broad portfolio right of activities. And this is just an example of some of the use cases under active development with or ecosystem partners. And you see here a broad set of categories. But one of the things that really comes to mind In addition, at the bottom, for example, you see things for the financial sector, like race come financial transactions, settlement, and so on. One category that really comes front and center is these world of applying quantum computing, to the world of chemistry to the world of materials. And, you know, I'm inviting you to be on stage to share with the audience. Why is that so relevant? And why is quantum going to impact our world?

 
Jamie Garcia 

All right. So why?

 
Jamie Garcia 

In 1981 38 years ago, IBM and MIT co hosted a conference on the physics of computation. And you may recognize some of these people. There's some pretty famous people in this photo, one One of which is Richard Fineman, Fineman, Frank famously said at this conference, this quote, nature isn't classical Damn it. And if you want to make a simulation of nature, you'd better make it quantum mechanical. And by golly, it's a wonderful problem, because it doesn't look so easy. And boys that true. So here's the idea. This is why chemistry, this is why we're interested in chemistry. So going from a design idea to something that can be commercialized. Something that can be a product is a multi step process. A lot of the bottleneck comes down to making experimental observations, and then having to explain them which we use computer simulation for. There's a back and forth that goes between lab testing and computer simulation. And this can take months, years and a ton of money to actually be able to figure out what's going on inside the flask before it can go on to scaling. In order to simplify this process, we've developed a number of Different ways from a classical perspective to actually look at chemical systems. And they range the gamut between chemistry and materials and start falling over into the world of engineering as we go to increasing link scale, so as we go from something on the atomistic level to something on the me so scale, it goes that the accuracy decreases as you go across. And so what we've started looking at with quantum computers is quantum chemistry. So quantum chemistry is understanding the very most fundamental behavior of molecules on an atom scale. So this is essentially the bond forming bond breaking reactions, and how and describes how chemicals react. If we are to better understand this, we can then expand that to understand things like ground state properties, so different energies of the molecules to be able to compare them to each other. That'll enable us to understand chemical reactivity potential. Energy surfaces, we can understand properties of bulk materials, and then eventually be able to apply it to things like telesis, which arguably impact all areas and chemistry. So what why he's quantum computers. So this is the basic concept, classical algorithms to be able to accurately simulate a molecule scale exponentially in size. This means that every new atom that you put into the molecule every new electron you put into the molecule, that math scales exponentially when it comes to modeling it on a quantum computer. This is essentially why we have to go to more approximate methods as we go across that plot that I showed you. However, quantum algorithms scale polynomially in size, this means that we can achieve highly accurate calculations and simulations for molecules while using much less compute. And that essentially means it will save time and in that back and forth, time is money. And we We've actually done this. So we've started looking at the dissociation profiles. That means when two atoms are pulled apart and the energies of the atoms as you pull them out of the system as you pull them apart from each other, we've been able to show that we can do this with an exact energy surface, and then actually run it on our quantum hardware. And finally, chemistry applies to all sorts of different industries, agriculture, aerospace, transportation, pharmaceuticals, you name it. And you can think about the potential power to be able to harness quantum computing to be able to use this towards looking at some of the world's biggest problems for industry, such as nice nitrogen fixation, new catalysts for co2 conversion, high performance polymers to lightweight airplanes, antibiotics, electronic materials and electrolytes for potential next generation batteries, and the list goes on.

 
Dario Gil 

Thank you, Jamie. And so one of the one of the A key observations that you brought forward is, you're taking something that is exponential in the number of calculations you would have to do with a classical computer. And you map it to something that is polynomial in how you perform that computation. And remember, exponential things don't scale well. Once you have an exponential number of variables, it's not a question that will solve it. When you have a more powerful classical computer. It's just impossible to solve. So I think that that's the most fundamental thing to realize that they were problems that we've sort of put under the rug. Because we just couldn't tackle them. The best we could do is to approximate Yeah, and in this world of chemistry, that equation is changing, right? Yeah. So thank you so much. Thank you. And let me now invite on stage Andreas Hintennach. So I'm going to invite two key partners and they asked please come in from Vijay Swarup, from Exxon Mobil. Thank you Welcome, who are two key partners of the IBM q network and who have been pioneers in exploring some of the implications of quantum computing for their industry? So let me pass this on to you Vijay for a second. But let me begin with the first question, which is, could you share with us in the context of both Exxon and the energy industry? What is the perspective that you're seeing in terms of like, what are the greatest opportunities in the decade ahead? And some of the biggest challenges? And so I'll ask you the same thing in the context of Daimler on the role of automotive.

 
Vijay Swarup 

Well, thanks for the question. Dario. Thank you for inviting me up on stage. And thanks to all of you for coming out. Today, on the end of what I'm sure is a very busy day for all of you. You're very much, Dario, to what you said about the the dawning of the quantum and the fact that we have the next stage of computing I want to talk about developing the energy future because we're actually at a similar point in energy where the challenges of energy are profound, and their game changing. And we need new solutions and the solutions are going to be deep in technology. And so as I go through this, as the scene said, I'm just going to talk about the, the state of play today and energy, what the challenges are and how we're addressing the challenges. And then I'm going to bridge into some of what Dario and we just heard around Jamie talked about in terms of well, you know, when you look to Jamie's last slide, of all those industries, every one of those industries, is connected to energy. Every one of those is underpinned by energy. In fact, everything in this room is either a direct descendant of energy or a first derivative of energy. It's something we take for granted. But it is there. And energy is equal to quality of life. And that is actually the challenge. So energy equals quality of life. This is a very simple chart that shows the UN Human Development Index, the Human Development Index takes into account things like living standards. Health Education, things like that. And then on the x axis, you can see the energy use per capita. So it's not surprising that as you go to the more developed regions, the energy is goes up. It's simple as you as you go higher in Human Development Index you get you do things like lights, like washing machines, like cars, things like that. And so you can see the big difference between the developed nations and the developing nations. And of course, the goal is to get everybody to the upper right. The goal is to get everybody to have the quality of life that most of us in this room, enjoy. And yet there's a billion people today that don't actually even know what I'm talking about. They don't even understand that these options are out there, there's the lack of energy, and they want that gap closed. And as I alluded, the, the challenge is really in the non OECD or in the developing in the developing world. So here you can see the co2 emissions that come from the developing world out to 2040 versus the developed That is the dual challenge. The challenge is how do you get the energy that people want, while addressing the co2 emissions? That requires improvements on what we're doing today, we call those efficiency. And that inquiry that requires new technologies, and whether its batteries, whether its biofuels, things like that, which you have to remember about our industry is it's deep. It's deeply science. It's deeply technology. And I'll talk about that more in a minute. But if you look at this problem, it started with a simple problem. I want to energy energy equals quality of life. We've now broken into two part problem which you have a challenge in developing regions, which is how do you get more efficient you have a challenge in developing regions developed, more efficient developing, how do we do it while reducing emissions, but it's actually more complicated than that because you really need to break it into sectors. As you can see here, you have transportation, you have power, you have industrial Then buildings is kind of a somewhere a combination of power. It's really a subset of power. So it's really three key sectors that drive the emissions. And each one of these is going to require unique solutions. So we took what sounds like a one equation, one unknown, and now we're already up to at least six equations. And if you break that down into Well, you know, you've got all sorts of things by country by state by region, it becomes a very complicated problem. So the question is, well, how do you take on such a hard problem? And the answer is technology. We have to look at technology solutions. We at the, in essence, are a technical company. I'm sharing the stage here with two other companies that are deeply entrenched, deeply, deeply technical at its core. And so how do we do this? So how do we develop affordable, scalable, lower carbon technologies? Well, the first thing is, you have to start with your own team. So we consider ourselves a technical company. We have strong internal technical team village you can see some of the numbers there. 20,000 engineers, 2500 PhDs, many members of National Academy, the Nobel Prize was just awarded to lithium ion battery. lithium ion battery, of course, is the core technology that allows a lot of the things we enjoy today from laptops and cell phones, etc. that invention was, was at Exxon. And then we've had over 250 patents. And we also have a consistency in r&d, which is really important. You've got to have a consistent approach to r&d. Now, as good as we are, we know that we don't have all the skills and to take on these types of challenges. One of the keys here is collaboration. When you're trying to optimize, you can do that singularly, when you're trying to invent when you're trying to disrupt or into trying to change, collaboration is needed. And so we collaborate across a broad range right from universities, energy centers like MIT, Princeton, Singapore, Stanford. National Labs, the you can see any TL and n ro, and then companies like IBM.

 
Vijay Swarup 

And as I, as I wrap up here, I just want to point out that one of the core things to energy is it is foundational. And what we've been able to advance in energy has been computational capabilities, from simulating reservoirs, to understanding seismic interpretation to optimizing grids. This has all been enabled by tremendous advances of computers over the past several decades. And so quantum we're looking for if we look into the future, quantum computing, should be able to address and open up even more opportunity. So we're excited about the future. We understand the challenge. It's a technology challenge at its core. And our approach is to work through collaboration and collaborations like IBM.

 
Dario Gil 

Thank you. Thank you. So Andreas, I wonder if I could ask a similar thing a perspective right, as you look at the decade ahead. From a dialogue, you know, point of view and not a motive for what's coming, what are the challenges and opportunities first?

 
Andreas Hintennach 

Okay, thank you very much Dario for the invitation. Thank you for all you listening and attending so far, there are major challenges actually. And which I already pointed out that energy is one of the core problems, or the four core challenges that we have been facing since decades and are still facing. If you look here at CES, what happened to Consumer Electronics Show, it started with micro computers and small devices as computer consumer electronic is and bar now actually became a major competitor to the automobile conferences like Detroit Motor Show and the others. This tells actually two stories first, maybe Consumer Electronics is maybe not as exciting as enough so that you can invite others I doubt that this answer is the right one. It's exactly the opposite. We kind of three was facing tremendous challenges after 130 years when our founding fathers invented the automobile actually gets into consumer electronic industry. Back that days, I have no doubt that horses had a way to talk to each other can kind of communicate to each other, but I can promise you nowadays cars can are more in the digital way. Nevertheless, the kindest way in the whole way legal industry is fully transforming. I would like to give you a rough example, for about 130 years innovation was just the optimization of one core invention that was the internal combustion engine that car industry or vehicle industry generally used. Now and actually we wish I and we share the same passion actually getting carbon dioxide emission to a lower level for in particularly cars, that is one option to electrify them or make them full electric vehicle. That implies that the old mechanical industry that has grown for more than 130 years, more or less flips into an electrical company. Money and if you think about how our bedroom is set up even in a software company to give you a rough flavor, but the bedroom is about is about three to five times the computational power of a desktop computer a personal computer and about two to three times the data that is processed in terms the program code the Microsoft Windows or the Macintosh operating system, the so called battery management system more or less flipz our company also in a kind of a software company enabling new technologies like sensors and as well as energy conversion because otherwise we will never ever face the the idea how we can mitigate the issues that some hydrocarbons cat will never burned down in a combustion engine. So electrification solves the core problem that is also had been associated with transportation, that is carbon dioxide emission and gifts of course, some freedom back to the driver. Cars become also autonomous and Something that had been associated more than 10 years ago with the Consumer Electronics Show, I have to admit that 10 years ago it looked around, it was almost around the corner and it's still not yet there. But I can promise you it will come. It's still around the corner, maybe a little further down the road, but it will come it will start in some of the segments in the market. and everything like that shares a lot. There's a lot of technology needed. And it's extremely important for even long term growing industries like ours, man, Mercedes Benz to get started with technology and invention as soon as possible, or as early as possible to be prepared once the change really happens. And if you think back, back in the day when horses were present, no one asked was asking for combustion engine that people were more or less happy with the horses. Well, sometimes they had their own mind, but I can promise you the first combustion engine even they had their kind of their own mind. They were not as reliable. But then a inventors came up, brought a new ideas looked at very crazy. No one was But likely to founded, even the king of the country didn't believe in the technology because it was believing in the horses. But I remember some people even in IT industry said, No one needs more than megabyte of core memory. And that brings us everything back into sometimes you have to believe in new ideas, believe that you can really do it. Don't wait for technology, let's say like, like the electrical grid. Don't wait for governments to found something like hydrogen filling stations, you just have to be present in the market. That's one of the core challenges that we are facing. And we are the strong partners in the world in particular universities, but also with IBM, a strong partner for computational industry like quantum computing that we will talk about in the upcoming discussion against and that's truly a very positive perspective for the future because we are just opening the door into a fully new field of technology. Thanks a lot. Thank you.

 
Dario Gil 

Thank you. Thank you. So what, you know, when you were both talking, just an observation that that just came across? If you look at the evolution of our respective industries, I'm not gonna say we were coasting. But there was a very well defined vector of what progress look like. I mean, in energy video, you shared with me that it was all about efficiency, right, with which we could execute the entire energy process, right, from extraction to delivery and so on. I mean, in the world of the combustion engine, right, for many decades of improving the efficiency but of unknown paradigm and, and the world of computing was Moore's Law, right? Where's the world of bits, making them cheaper, and, you know, more broadly available? And I think what we all have in common in the different industries, is that there's a fundamental big difference that is happening in this decade, that the decade ahead is not just about turning the crank 10 more times of the same, whether we're talking about triplication are in the challenge of now decarbonisation or in the computing industry, adding the entire neural architectures for AI and quantum computing as a core paradigm. So I wonder if you could address this aspect of the urgency of change. And whether you perceive that there is a technology gap, because it's not easy if you've been innovated in a particular way for a long time. Now, all of a sudden, you've got to retool the company, you got to do a lot of all the things to change the vector, especially at the scaling which we all operate. And I wonder if you could address whether you perceive a technology gap. Where do you see that you need to make investments to close that technology gap, and very specifically the role that quantum is going to play in closing that technology gap in the decade ahead? So maybe I'll be Jamie will start with you on

 
Vijay Swarup 

because I think some of the answer is in your question. So let's start with is there a technology gap? I think that it's somewhat self evident. If there wasn't a technology gap, we wouldn't be facing the challenge that we're facing right now. So while there are technologies available, they clearly aren't sufficient to meet the demands of a growing population with a need to reduce co2 emissions. And so we need different technologies. Now, it just so happens and what you what you said without saying it is what we would we call parallelism. So, it just so happens that while we're on this need for breakthrough energy solutions, you and others are coming up with breakthrough computational capabilities. And of course, a lot of what we're doing today was enabled by the breakthroughs in computational capabilities over the last several decades. Our ability to be as efficient as we are today, is because we can simulate and we can model to the extent that we can model today whether it's computational chemistry, computational Physics, which are the underpinnings of what we do in energy. So whether it's separations, whether it's production of energy, consumption of energy, it, you go back to basics. It's chemistry, it's physics. But it's also, as Jamie and you also mentioned Daario. It is an approximation. And we are hitting some of the limits that today's computers can do. Now I'll give you a real quick example. Some of you know that we have a pretty big biofuels program. And the breakthroughs we've had in biofuels have actually been enabled by two adjacent technologies. The first is gene editing, the CRISPR casts and things like that, which comes out of the biology sector, and the other is the speed of the computer. So it used to take months to sequence the genome can now be done in weeks. Well, part of solving this biofield challenge is a little bit of trial and error. And the faster you can sequence and the more accurately you can sequence and understand what the various parts of the organism are doing, the faster you can get the Oregon Introduce the oil, the hydrocarbon the oil that we need to move the trucks and the airplanes. Well just imagine if if the stuff that Jamie showed is plays out the speed at which you can do the computing the accuracy by which you can do the computing could take us to the next level on the material side, which is the core component of a battery. At the end of the day, it is a materials challenge. And we're limited to where we're making approximations. Well, if we want to get to the next breakthroughs in that chemistry, we're going to have have to have better computational understanding quantum you know, people think a quantum will never quantum is core. We all took quantum chemistry and quantum physics Okay, chemistry and physics are core to energy, quantum chemistry, quantum physics. So it comes together and it is a it's it's it's a nice thing that these things are being developed at the same time that the challenge we have and what we're working with IBM and is how do we start working these things anticipating the the problems that can be solved with this With the quantum computer, and how do we use those, so that we can advance to get into the solution to getting the options identified quicker. So we have a gap, we have a concept of how to close the gap. We know it's going to take science, we know it's going to take technology. But you know, it's also going to take patience. And if that's the one thing that we have to remember, as we go down this, this pathway, which is there is going to be hits and misses. Doing things in parallel, trying to do things in the most efficient way is really important.

 
Dario Gil 

But that's why it's so important to be grounded in the right fundamental ground

 
Vijay Swarup 

and the right fundamentals, and then collaborating with the right people. That's right. So we're really excited to be the first energy company that joined the IBM q network. We wanted to get it on the ground floor, we wanted to understand how it fits into the energy space and we want to work with with the best. And IBM has we have a long history of working together on innovations and energy. And we see this as the next chapter in our collaboration. Thank you.

 
Dario Gil 

Thank you. And that's what is your perspective on these technology gap and whether exist and the role of quantum and Closing?

 
Andreas Hintennach 

Absolutely, yeah, absolutely. I agree there is a gap. I would even go to the question and the answer that there are multiple gaps. And I will come back to that comment in a second. You're which I already pointed the discussion to the battery battery consists of basically chemistry. back those days, when there were only combustion engine, everything was associated to carbon and hydrogen. So hydrocarbons as a chemistry, I can promise you carbon is just one atom, but it's not boring. But if a battery opens the whole field of the periodic table of elements, because it literally stores the energy and converts energy in a very different way. So we have to play with a high complexity of what nature offers you. Usually once you have a large potential of different elements to play with. On the one hand, that's very beautiful because we have more options, more opportunities and chemistry. On the other hand, it gets so difficult because you don't have to run maybe, let's say literally spoken 100 experiments, maybe 100 1000 and we don't have another hundred thousand years of evolution to wait that nature more or less a nature's a very efficient experimentation or does it for us in a way just to find it out and optimizes it over the years. And actually we if you follow the carbon dioxide climate discussion, we don't even have another century to wait for this illusion that it somehow falls from the sky and we find it. And finally and that is why I made the comment maybe we have multiple gaps. In the end, there are multiple passes how to tackle that, which had already pointed out the biofuels. There's also hydrogen and the fuel cell but in particular, also the battery. So it may be there are different ways to tackle the holistic question how we increase sustainability. And that is another kind of a gap in science. We have no standard so far, almost no standard, how we actually describe holistic sustainability. You might know recycling and some other programs from your daily life. But that's an old image Goal because as you have seen in the previous slides, energy is one of the core issues. It's about preserving nature preserving some kind of how we do technology, but also how we do education. If we don't educate the next generation, that's not sustainable in a way, how we actually form the next generation of material scientists, engineers, and people thinking about problems first, and then of course solution to those problems. This is by only working with the best of a network. And the US and a lot of other countries in the world provide excellent research institutions as your company dabio to partner up with and really work together. And when I had mentioned before, 2020 not only opens the door into a new decade of a lot of opportunities from chemistry and physics, I have no doubt that it also opens the door in the new world how we need to interact with each other. When the founding fathers The two were two guys that invented the cars they have never actually met that even invented it in parallel because they didn't know each other thought geographical distance those days was not very high. But nowadays actually there is no genius in the world that has an answer or potential answer to potential questions that we need to answer that solve some of fundamental problems in energy. And actually walking individually on some innovations, at least to my understanding, won't walk very well. So 2020 and 2030 will be decades of a different way to collaborate, maybe even very different ways, how academia, different types of industry need to interact with each other, to holistically start to solve the basic problem, like sustainability and in particular, the carbon dioxide problem. And if we go back what quantum computing really can help that is a completely new technology. You had shown that nice picture back the state's been programming a computer was playing with relays and mechanical stuff, actually, and then the back was really abuk literally spoken for mechanical industry, people had to collaborate over the scientific fields, there were no computer scientists back those days was electrical engineers, physicists and so on. Actually, we can learn from what had already sped up the computer science field. And for us this this is why partnering up with your company w IBM actually brings us an excellent partner and in particularly the core that we cannot do as an industry. We cannot do the operating system that you had pointed out before of a quantum computer, we cannot do the hardware research that your company does. This is why it is a good example in which our collaboration is the best way to speed something up but also to act as an enabler. Yeah. So on that very topic of collaboration, you both were exemplars of being leaders within your company have seen something ahead of others. And then using your capital in the company to say we've got export quantum computing, I'm sure at the beginning was an easy it's like what the heck is this about? Right a few years ago as you were thinking about it, you know, this is so early where we mean, although you did understand the relationship between computation and value, but what is this new? So I wonder if you know, perhaps are others listening to us or in the audience who are in a similar situation where they understand information and computation and the power that it has to their business or in their institution, but this is a new area quantum. So how do you decide to do it? And what do you have to overcome? And what do you have to create, like what team to get started, what approach that perhaps you can share that, you know, has been productive to you as you ramped up their efforts. So, actually, the first idea was similar to what you had pointed out, it came from quantum mechanics, so more or less the operating system of chemistry or of nature, because then the closest way actually is to think about quantum physics and how potentially a new IT system might work. They were facing the situation that traditional silicon based digital computers were at the end of their computational power, and even Waiting another 18 months to buy Moore's law to wait for double the computational power was just too long, because that had meant meaning maybe waiting for another one and a half decades to get the results we really needed. So it was ready in mid 2015 1414 ish, whatever it was to think about really would will all be able and are we willing to get into a completely new field that has never been the core of our industry? And that had pointed out before we already were on our way into getting also software company. But to be honest, it was a crazy decision. It was hard to make that decision because there were not very many people. Maybe you can believe me, in our company that had an idea what quantum computing might be about. I agree to the fact every student someone hears about quantum mechanics. I'm not sure how many people love that topic. Some will. Some won't. It's Up to up to single person but it was a tough decision that we made the decision that we need also need to build up resources. Because counting at universities, you want a chance to find a computer scientists being able to do something with quantum computing. So it was really building our people. That's how we tackled the problem. The slice the problem down to LinkedIn to use cases. We call it a use case. For example, if we look for a particular chemical compound for a battery on catalyst for fuel cell, or another option like a logistics problem way we optimize the GPS routing in a car. And then they were looking for potential partners in IBM is a strong supply of us of our because of our mainframe and IT industry. And then it was certainly was to ask a question first, do you know anything about quantum computing? And we were very, very happy actually, that we found partners to be able to talk to that understood us and health doesn't leverage that problem. And if you just think back that is just three years ago. Look at the progress of what has happened in three years since then. I mean, it's really a spectacular situation. And the rate of progress is accelerating. So I mean, it's a well placed bed, but early on, it wasn't so easy. So yeah, maybe add, let me add one just funny note, there were a lot of problems. And you will always find that in your organization that they'll tell you from day one, it will never ever walk you based your lifetime, you waste your money, and it was so nice to prove them wrong. I love it. That's great. So how about for you?

 
Vijay Swarup 

Well, as you know, a year ago, we were here announcing that we were the first energy company to join the the IBM q network, and we were very pleased with the progress. I think, you know, I'm going to go back to some of the points I've already made. First off, it is understanding that we have a technology gap. Number two is understanding the very nature and the value of computational capabilities that underpin a lot of what we do in energy. And then number three, it's looking for the right collaborators. So do we have the complementary skills bureau expertise in hardware, your expertise and and we have a history of going back from reservoir simulation to computational chemistry to being able to simulate and understand the molecules that make up a barrel of oil. All of those are based on lots and lots of collaboration between our companies and Andreas point, we, we know, we need new paradigms, we need paradigm shifts. And so this just came together at the right time where we're trying to look at the next generation technologies, the next generation of materials and the capabilities around quantum we think can really really help there to how do you get started? I think it's a lot of what Andrea says firstly have the right capability. So we bring in our computational chemists, and our computational physicists and folks like that. You bring in your your core capabilities, we get the right all star team built. And then we start with we start slowly and we build from there. And of course, we had a review back in December, we saw the progress we've made in a year. We're very excited. Again, we're convinced that getting in on the ground floor and shaping how this evening is the way to do it. And this is one where, you know, it's really it's an inclusive approach, because we're looking for the right people the right skills to be able to

 
Dario Gil 

based on this idea of collaboration based on the idea of collaboration, integral important. So let me you know, now in the last minute, take an opportunity to close some thoughts. You remember early on, you know, when I began the presentation, I talked about this perspective about how the future of computing was the simple equation of bits plus neurons plus cubits coming together, and the me offer even in the world of materials discovery, and accelerating science, that if you look at three distinct constituencies, the world of bits and classical computing and high performance computing, this decades of work of using these classical computers to be able to do computational modeling, right. If you look at the world of AI and neural based architectures today, some of the most exciting work happening right now in the area of discovery also not come from a new methodological approach that is data driven, to be able, not starting from the core equations of physics, but a data driven approach to be able to create predictions. And if you look at the world of quantum, as we Jamie showed with a quote from Richard Fineman, the very idea of exploring quantum computing was in the context of modeling nature. So here's the thought I want to leave you with three independent communities, each one, making significant progress in pushing towards this frontier of accelerating science and discovery in this context of materials. as impressive as the capabilities of each of those communities is going to be the will of AI the wall of quantum and the world of classical computing, to accelerate discovery. Imagine what is going to happen when this convergence of all three When we have computing architectures that bring the best of bits and neurons and cumulates together to accelerate this, things that would have taken us literally decades, we will be able to do in years, and things that would have taken hundreds of millions or billions to do. Can we do it in 10s of millions of less the implication for some of the most fundamental problems that we have in society, that in the end, they have a physical characteristic. That's why one of the things that unites us is like the world of materials is so fundamental to the world is going to be radically outer, as this computing revolution unfolds. So I'm very excited about the opportunities that this will have to have a such a meaningful impact to the world. And I will just close on this idea of partnership. that technology is great, but it is the combined expertise of so many institutions that have to come together to be able to do this to tackle these companies. problems. And thank you visa and undress for joining me on stage and thank you for your attendance today.

CTATECH-PROD1