A quantum computer that's checked by Big Tech? It takes 30 years to commercialize.

■ Broadcast: YTN Radio FM 94.5 (09:00-10:00)
■ Host: Reporter Cho Tae-hyun
■ Air date: January 15, 2025 (Wednesday)
■ Talk: Reporter Oh So-young of The Guru

* The text below may differ from the actual broadcast content, so please check the broadcast for more accurate information.
◇Cho Tae-hyun: We've talked about quantum computers before. I'll go into more detail about this. I'll connect you to reporter Oh So-young from Douguru. Is the reporter here?

◆Oh So-young: Yes, hello.

◇ Cho Tae-hyun: Quantum computer, I think it's a word that you've seen a lot in newspapers and broadcasting this week. I can't help but talk about CEO Jensen Hwang. Why is it related?

◆ So-young Oh: Yes, Nvidia CEO Jensen Huang was recently asked about the timing of commercializing quantum computers at a CES meeting in Las Vegas. So many people said they would believe it if they said it was about 20 years, but in fact, you have to wait 20 years for a useful quantum computer to come out. After talking about it like this, there is a situation where people are saying that it's right or wrong about this.

◇ Cho Tae-hyun: So expectations for quantum computers are high, but shouldn't we wait about 20 years for this to happen? Talking about this, it's affecting the market and various areas. Quantum computers are a technology area that is receiving a lot of attention right now. Please explain easily what a quantum computer is.

◆So-young Oh: If I can explain it compared to the existing computer we know. When processing the data values or data values we input, general computers only understand and process them with two numbers, 0 and 1. So, for example, if we type the video in text like this, the general computer processes it like 00, 10, 01, 11. So we call the smallest unit of information that is represented by these zeros and ones bits, but quantum computers use qubits rather than bits.

◇ Cho Tae-hyun: What is a qubit?

◆Oh So-young: The biggest feature of the qubit is the superposition state.

◇ Cho Tae-hyun: So there are 0 and 1 at the same time.

◆ Oh So-young: Yes, a state where 0 and 1 are simultaneously is called a superposition state. So in quantum computers, the smallest basic unit of information becomes qubits. Three qubits can process two times two times two times eight information at the same time, rather than three. That's why people say that the calculation speed is faster than that of existing computers.

◇ Cho Tae-hyun: I think this part is very hard to understand, but it's hard to understand because it means that the front and back of the coin are at the same time. I think it's better to just accept it because both of them have that part. How fast is this quantum computer compared to conventional computers?

◆Oh So-young: Compared to supercomputers, quantum computers can solve a problem that takes 1 billion years in 100 seconds. So, there are expectations that you can solve complex codes or solve difficulties in the fields of chemistry and bio, but it's hard to assume that it's fast in all situations. So, the field that quantum computers are good at is that quantum computers are very fast when learning about simulations or optimization models that need to improve various possibilities. For example, when I go to a broadcasting station, if there are about 10 paths, the existing computer goes one by one, calculates the time, and shows it as a whole. On the other hand, quantum computers start at the same time and show one of the fastest paths. So it's bound to be much faster.

◇ Cho Tae-hyun: These things can be seen as the advantages of superposition. Then the general public bought a quantum computer this time, it's not going to happen much and they will mainly use it in certain fields. Can I look at it like this?

◆Oh So-young: Yes, there's a lot of talk about encryption right now. And I think it will be applied a lot in the industrial field in bio and finance. I think it will take a long time and be difficult for the general public to use it, as if there is no supercomputer in the general public's home.

◇ Cho Tae-hyun: Well, as you said, I don't think there will be a case where a supercomputer is put in your home in the future, so I think it will be. The concept of quantum computers appeared a long time ago. But recently, it's receiving a lot of attention. Can I take this to mean that there has been that much progress?

◆ So-young Oh: Yes, first of all, there was progress in terms of research. One of the important studies is that Google proved its quantum advantage in 2019. This was also published in Nature, and it proved quantum superiority. What does this mean? It may be a little difficult, but you can say that it proved the possibility of quantum computers. So at that time, using quantum computer Seacore, supercomputers performed calculations that took 10,000 years in 200 seconds. So it proved that there is potential for quantum. It proved quantum superiority. Some of the research results came out like this. And there is progress in the research results you mentioned, but AI is receiving a lot of attention right now. So you can also connect with AI and think about it. In the end, in order for AI to develop further, it needs to be able to perform high-level operations and complex simulations, and there is an expectation in the market that these things can be more helpful if quantum computers are commercialized.

◇ Cho Tae-hyun: So the possibilities of helping each other and developing quickly are reflected to some extent. But if you listen to CEO Jensen Hwang, it will take about 20 years. How does the industry see it?

◆ So-young Oh: The reason Jensen Huang said it takes 20 years is because quantum computers have a number of challenges. First of all, the qubit I mentioned earlier is very vulnerable and sensitive to external environments such as temperature change and noise. So, in order to keep it stable, it is necessary to maintain an absolute temperature, or a cryogenic temperature close to minus 273 degrees Celsius. So there's also a question of whether it can be maintained stably. And as the qubit increases, errors in calculations and operations occur. So the problem of correcting this error is still a challenge. However, the key is how much earlier this difficulty can be brought forward, but this is also different from the timing of quantum computing companies and Jensen Hwang. So, one of the companies that has been talked about a lot recently is IonQ, and in the case of Duke University professor Kim Jung-sung, who is a co-founder here, Jensen Hwang said for up to 30 years, and this 30 years means that there are 30 years for such a large corporate quantum computing company like Nvidia. He reinterpreted it again. And in the case of Google, I thought that it would take about 10 years to commercialize quantum computers at the end of last year, that is, about 10 years to actually use them. So there is still some debate about the timing of the commercialization of this computer.

◇ Cho Tae-hyun: It may not be 20 years, but it will take a lot of time and it may take longer than you think. But the reason I said up to 30 years, this was a way to keep quantum computer companies in check. I think there's also this evaluation. Why is this kind of assessment coming out?

◆ Oh So-young: Yes, it's because of the graphics processing unit that GPU Nvidia knows best. The reason why this graphics processing device has received attention so far is its excellent computational function, especially in the AI era. It is possible to process vast amounts of data quickly, but when this moves into the era of quantum computers, the story is changing more about whether GPUs can gain the upper hand. So, considering that quantum computers have shown faster abilities than existing computers, quantum computers are bound to have an advantage in terms of speed. If that happens, GPUs could fall behind. With this talk, there is speculation that CEO Jensen Hwang was a little concerned about this and predicted the timing of the commercialization of quantum computers late.

◇ Cho Tae-hyun: I think there's an analysis that might have been a little bit of a check. There are several companies that develop quantum computer technology. There are Google ahead, IBM, and several companies, but there seem to be differences in direction. What's the difference?

◆ Oh So-young: Yes, let's increase the qubit to make a lot of computations faster, and let's reduce errors. This goal is the same. But I think there is a difference in how to make qubits and how to tolerate errors. First of all, if I talk about the error rate, IBM says that it will make a program to remove noise that causes errors because it wants to prevent errors in advance. And in the case of Google, I will make a program that manages errors in real time. That's what we're talking about. Therefore, a combination of qubits into one qubit is called a logical qubit, and the error correction system is operated using a logical qubit, and the actual research results have been published. And I told you how to make the qubit that I mentioned earlier. This part is different, too. There are two ways. There are two methods, ion trap and superconductivity, and I'm sure you've heard a lot about superconductivity. Making it into a sudden low temperature and reducing electrical resistance to zero is called a superconducting phenomenon, and this phenomenon is used. So, the superconducting method that uses superconductor circuits to create qubits is mainly used by big tech companies such as Google and IBM that we know. And in the case of ion traps, you can use the electromagnetic field of ions to see it as trap powder. So using this to make qubits is IonQ, so the ion trap used by IonQ is a little stable, but it is shown to be relatively slow to generate qubits rather than superconducting methods. On the contrary, superconductivity has the advantage of being able to produce qubits on a large scale very quickly.

◇ Cho Tae-hyun: I think the people who are listening now will have a lot of headache. Korea is a little late in the quantum computer world. In any case, it is attracting a lot of attention as a future industry and seems to be needed, so I think the government needs a lot of support. So far, we've talked about quantum computers with reporter Oh So-young from Douguru. Thank you for talking today.

◆Oh So-young: Yes, thank you.

#Jensen Huang #Zuckerberg #QuantumCom #QuantumDynamics #TechnicalStates #Qubit #IonQ #QuantumOverlapping #Superconductor #IonTrap #Commercialization