Thursday, September 29

“Few people are going to have a quantum computer and most will not even need to use it”

In 1984, Charles Bennett and Gilles Brassard invented quantum cryptography, a system that guarantees the inviolability of communications. “Quantum information is like dreams, they cannot be copied or shared,” Bennett compared after receiving the Frontiers of Knowledge Award in the Basic Sciences category with Brassard. They were not alone: ​​the mathematician Peter Shor, who demonstrated that quantum computers would jeopardize all current cybersecurity, was also with them. He discovered the ‘nightmare’ part of the quantum world once it actually lands on our world. “But it will not have the ‘superpowers’ that some grant it,” he reassured from the BBVA Foundation headquarters in Bilbao, hours before picking up his

award last September, while chatting with ABC.

– What is it like to receive an award like this?

-It’s very nice, of course. I feel honored. Even more so when I share it with Bennet and Brassard, who have been friends for a long time.

-Since the so-called quantum supremacy reached by Google, later claimed by China and passed through the flour of controversy. Is Quantum Computing Hot?

“Yes, of course,” he laughs. I do not know if it is something temporary, although it is true that it is necessary to clarify that quantum computers will not do many of the things that some expect. They can be very useful for tasks such as saving computational time for pharmaceutical companies trying to simulate quantum mechanical calculations with classical computers to create models that advance how molecules and drugs will behave. Carrying out all these operations on a classic computer is very inefficient; but if you have a quantum computer, it is much simpler.

– Several organisms already speak of quantum computers of 50, 60 and even more than a hundred qubits, the quantum bits or the ‘language’ of these processors. In what phase to ‘feel’ a functional quantum computer are we?

-In a very early stage. In ten years we will probably reach 1,000 or more qubits, but it will still not be enough to carry out all these applications that are being promised, although some useful things can be done. We probably need 10,000 to a million qubits to have really efficient systems.

-Now one of the big problems is the correction of errors. The qubits have more and more quality, but for them to really work they have to have a very low margin of error …

-Yes it’s correct. A couple of bug fixing programs have been run and they work fine. But still the margin of failure is very wide. The hope is to make better and better qubits with fewer errors, and to create many of them, since the theory says that the more qubits you put into the system, the fewer errors it will have.

-Its algorithm compromises all the classic encryption. So said, it’s a bit scary.

-Can be. However, there is another encryption algorithm called error learning that does seem to resist quantum algorithms. The problem is to replace all the codes with new ones that guarantee that security.

-He did math. How did it end up in quantum computing? It is a field that has barely three decades of development …

-Before graduating I had to choose between physics and mathematics. I probably didn’t choose physics because I didn’t want to do a mandatory experimental laboratory course that everyone was quite scared of (laughs). So I opted for math. Then I went to MIT and there I did a doctoral thesis, half mathematics, half quantum computing. Then I started working at Bell Labs, where I was able to bridge the two areas. In the 1980s, Bennet gave a talk there about his quantum key distribution system. Then Bennet and Brassard published their article and I was very interested in it. I started looking for references about it, but there were none. At the time, Umesh Vazirani – one of the founders of the field of quantum computing, whose article on the theory of quantum complexity defined a model of Turing quantum machines that was susceptible to complexity-based analysis, and the basis of Shor’s algorithm. – and gave us a talk on quantum computing. And since he had already taught physics classes, I understood very well what he was saying. That was in ’93. From there I started to think about what could really be done with quantum computing. I was at those when I saw Adam Simon’s algorithm, who had enough technique to perform a factoring algorithm.

-Still, quantum computing was in its infancy. Did you imagine its potential?

-I already knew that factoring was a major problem. And there was a lot of excitement about it. But there were also those who said that quantum computing would never be possible. For the errors, mainly. And then there is the Heisenberg uncertainty principle, which says that you can’t measure a quantum error without destroying it. If you try to identify the errors, you destroy the quantum state, therefore the reaction stops.

– But now it is being demonstrated that all this is possible. What’s more, classical computers themselves can mimic quantum systems.

-That’s how it is.

-And will quantum computers give us the revolution they promise?

-It will not be a revolution like the arrival of classical computing. Quantum computing will be very useful for some applications, for which classical computing does not reach, and in that field they will be very useful. But we will never have a quantum desktop. On the basis that quantum computers have to be in physics laboratories because they need very cold environments that can only be achieved in cameras with very expensive equipment. In addition, maintenance technicians are also needed to adjust all the equipment. Not everyone is going to have a quantum computer. And, in fact, most people are not going to need it. Not even your connection to the cloud. Yes there are more likely a few very efficient quantum computers that you can connect your desktop computer to and run an application. That is more possible.

-Although it is conservative in its predictions, yes that quantum computing can offer a disruptive change in chemistry, medicine …

-Yeah right. In fact, artificial intelligence and machine learning, for example, which has had a great development in recent years, has exceeded our expectations and we do not know how far it will go. If it came together with quantum computing, it would surely be a shock. However, we won’t know until we have a quantum computer.

Leave a Reply

Your email address will not be published.