— Tell me, is a 20-qubit quantum computer ready?
— As part of the roadmap for quantum computing, we have developed a 20-qubit quantum computer. We implemented it on the ion platform. We also have a 25-qubit computer on a nuclear platform. But the quality of operations is better on the ion platform. We have already said that over time — and this time is beginning to come — computer performance will be much more important than the number of qubits. The further you go, the more difficult it will be to compare different platforms. 100 atoms is better than 30 ions, no? And the answer lies in the fact that it depends on which 30 ions you have and which 100 atoms.
Approaches to evaluating the effectiveness of quantum computers are being developed in the world. For example, IBM has proposed such a metric as quantum volume, a certain combination of the number and quality of qubits. Not everyone agrees with it, not all, let's say, aspects are covered by one parameter, but this is an interesting parameter, and we are also starting to use it for ourselves.
Next year, of course, we will talk about the number of qubits, we are starting to show algorithms, the depth of algorithms, and these will be no less important characteristics.
— You said earlier that by the end of this year there will be a 100-qubit computer. Have your plans changed?
— We have plans from 50 to 100 qubits. 50 should be completed by the end of this year. Let's see if we can get more.
— A simple question: why do we need a quantum computer? Which areas of computing does it allow you to close better, more promising than existing machines? How is it better than existing silicon technologies?
— The simple answer to a simple question is that we don't know why we need a quantum computer. But not because we don't know anything, in fact, of course, we know the first applications of the quantum computer, we have the first algorithms. We don't have enough hardware power to show the usefulness of the algorithms, but they already exist. And they will help solve problems in logistics, in modeling molecules, for example, and so on and so on — where many different options need to be sorted out. This direction will definitely bring its successes. For example, the Rosatom State Corporation is already testing the operation of quantum algorithms in the tasks of developing new types of reactors and batteries.
When classical computers were created, it was believed that thousands of such computers would be enough for the whole planet, because there is a certain number of tasks and they will be solved. Yes, they built computers, solved the problems, but it turned out that while they were solving these problems, they discovered a whole horizon of new ones. In this context, we do not yet know why we need a quantum computer. When we begin to solve the problems that we see today, then, of course, we will see new horizons, and that's where something very interesting will open up. This is how many great scientists think — that as you start working, playing on this computer, running different things, different algorithms, you will understand what else it can be used for, right now changing our lives very significantly. However, it is already clear that it can be used to solve problems of modeling substances, chemical or physical. By the way, there is a strong suspicion that a powerful quantum computer will be built on silicon technology, oddly enough. Only the first samples are appearing now. Silicon technology is the most sophisticated technology possessed by mankind. And we can make billions, tens of billions of transistors. And we would like to have at least millions of qubits. And in general, it is clear that they could be implemented on a semiconductor, silicon platform. We don't know how to do it on such a large scale yet, but the first results have been obtained. Everyone is waiting for the rapid scaling to take place. If it goes into silicon technology, then, of course, we will all switch very much to this. This is the first moment. And the second point is the quantum computer, when you asked how it is better, it is radically better in some tasks, and in some it will never be better.
— Which ones are better radically?
— In tasks that require parallel computing. Due to the fact that a quantum computer is in a superposition of a large number of states, using a certain algorithm, we can process many, many threads at once. And this multivariance, multifactoriality, are the tasks that are good for a quantum computer. But, for example, sequential tasks will be much better solved by classical systems. And here it is better to consider a quantum computer as a coprocessor. Artificial intelligence chips are now being made that are capable of solving artificial intelligence tasks better than previous platforms. And in this sense, a quantum computer can be considered as the next add-on that will solve another class of problems. The only difference between all these steps is that inside a quantum computer, logic is arranged in a completely different way due to the properties of the quantum world. But from the consumer's point of view, it will seem to him: Well, we've started solving some other class of problems.
— By the way, from the consumer's point of view. Humanity, as you know, grows up in the game. The computer world is a game world. Is it possible to use quantum computer technology for the gaming industry, in your opinion?
— It depends on what kind of game it is. If you try to create a simulation of complex physics in it, then as you begin to model deeply, you may not have enough power. Why am I talking about complicated physics? A classical computer cannot calculate the behavior of even a small number of particles. And quantum — it itself has a similar structure, and therefore, of course, it is more, let's say, convenient for it to carry out such calculations. Is it necessary in games? We'll see.
— Is there a potential danger in using quantum technologies?
— The world has been developing very interestingly in recent decades. Technology is rapidly and very much becoming more complicated. The amount of information is growing catastrophically fast. Over the past five years, humanity has accumulated more information than in the entire previous history of mankind. On the other hand, the interfaces are simplified. You use phones and computers more intuitively, although the content there gets more complicated each time.
— Is technology supposed to develop a person, not serve him?
— This is the main question. In my opinion, of course, they should. Because technology is a tool for learning about the world. Cognition of the world is a searching mindset, not the other way around. On the contrary, this is a way of simplification: electrodes are connected to your head, brain areas are stimulated, for example, the pleasure center. And then why would you watch social media to have fun? You can simply get all dopamine addictions through electrodes or some kind of transcranial stimulation there. And life will be beautiful, meaningless and the same. That is, it would seem that we built, built and came to a state of death anyway, oddly enough. Death is not biological, but cultural, intellectual.
— Returning to the quantum computer. And what is its accuracy? As far as I know, it is still low, and depending on the set of qubits, it drops significantly.
— The accuracy of our computer is 95% on a two—qubit system. At the same time, progress in accuracy is very fast. A year and a half ago, we started with 60% — this is slightly higher than randomness - now it is 95%. And then, I think, we will quickly reach 98-99%. Then engineering work will begin on even larger arrays.
— Any progress is based not only on the intellectual insights of scientists, but also on tools. What do we have in terms of tools in the current difficult situation? In terms of access to technology, to materials? What is the situation in the development of quantum technologies?
— There are problems. Indeed, if we were talking three or four years ago, then we used 90% of imported equipment. And now we have received a powerful incentive to develop our own technologies. Of course, it's impossible to do everything ourselves, but we managed to launch some complex lasers, for example. We are talking about femtosecond lasers. Of course, the most difficult thing is everything related to microelectronics. To make a chip, you need to use a lot of elements, a lot of materials, a lot of equipment. If one element of the chain is knocked out of you, you will not receive the product. In recent years, the most powerful sanctions have been imposed against China, and not only in the field of microelectronics. They don't sell lasers to them anymore, and they don't sell fridges for superconducting quantum computers. And they are forced to quickly deploy their import substitution program. Of course, China has a strong advantage over us in this area. In general, there are technical solutions that were not sold to us — now they can be bought in China.
— Do we have any good work on creating software for quantum computers?
— There is a whole direction in Russia, several scientific groups that deal with quantum algorithms and make quantum software. In this direction, we are quite on par with world leaders. The theoretical work of domestic specialists is of a very good world level. And most importantly, we see that some foreign colleagues, who are also actively developing quantum computer systems, are lagging behind us.
— You are developing the LIFT Center for Neuroscientific Interdisciplinary Research in partnership with the FMBA. What does this center do and what results does it already have?
— The center was actually launched in 2023 with the support of Gazprombank. Laboratories are being built and six research groups are working. The experimental results of the center will be available at the end of 2024. We have created such a center in order to improve the person himself from the point of view of the nervous system — both peripheral and central. For example, to create perfect prostheses that will be sensitive, not just functional. This requires deeper interaction with the nervous system.
— Tell me, how do you feel about the latest event in this area — the introduction of a neuroimplant created by Elon Musk's engineers into one of the patients?
— A large number of experts say that Musk shows not particularly breakthrough research. But from a marketing point of view, of course, he is number one, and the fact that he draws the attention of the entire civilization to such issues is important. We set such goals for ourselves. I'm not ready to name specific dates for implant experiments right now, but in general, ways to communicate directly with the brain are one of the tasks in LIFT.
Andrey Reznichenko