INTERVIEWS AHEAD OF KRYNICA FORUM 2023

Piotr Ciepiela, EY: We are two, three years away from building a quantum computer

‘In quantum computing, we see a tool allowing us to achieve previously unattainable goals, including the cybersecurity context,’ says Piotr Ciepiela, EY Partner, a global leader in cybersecurity architecture and emerging technologies.

 
Does the world need a quantum computer? This question is, I guess, a bit philosophical…

It absolutely is philosophical, an even rhetorical and science-realistic, too. In contemporary world, the technology is both whimsical and mercurial. Any novelty unveiled to thunderous applause and with much fanfare can soon lose its value in a dizzying fit of depreciation.

Due to computers, this process picked up even more speed.

Yes. And at the same time, we are living in an almost insatiable world. These two modern-day tendencies make the civilisational progress unstoppable, no matter whether we consider a given technology to be needed or not. If a company chooses not to invest in a particular technology, giving lack of time for research and development as its reasons, another organisation will take up the challenge to gain competitive advantage. In contemporary times, this advantage is most often a result of tech innovation. That is why it seems to me there is no way out. The road to the future is a six-lane highway and everyone is just rushing ahead at breakneck speed. And since technology is more available and distributed than ever, the chances and opportunities are open to practically everyone.

Do such advanced technologies generate opportunities or rather threats to humanity? You can imagine that a quantum computer will make it possible to crack security measures and passwords quickly. On the other hand, it’s possible that its computing power will turn out to be useful in solving problems that humanity is facing.

The advances in quantum computing are opening up new vistas in such areas as modelling, processes, material analysis, logistics, or effective decision-making. In quantum computing, we see a tool allowing us to achieve previously unattainable goals, including the cybersecurity context. We’ll always identify more threats lurking around, because this is in our nature, but it doesn’t change the fact that we won’t stop the progress. Accordingly, it’s about always having the appropriate awareness of how we can use quantum computers, but also what dangers they bring.

Exactly…

On the one hand, quantum computers may be used to crack current cryptographic algorithms, which we are using absolutely everywhere, even if the cryptography processes are carried out away from the users’ awareness, under the hood. But on the other hand, we have post-quantum algorithms, which can ensure our safety against such attacks much better. So we’re making note of a threat and seeing a potential response to it at the same time. Apart from the security-related issues, quantum computing opens up immense opportunities in various domains. The truth is that these are not simply faster computers. They are built different, behave differently, need other algorithms and software.

What does it mean for quantum computers to behave differently?

A standard computer works following a bit logic, that is binary logic. A bit can remain in one of two states – zero or one, so either something is there or it isn’t. The system is used for programming computers, so as to make them able to multiply, divide, and calculate some values. When it comes to quantum computers, however, instead of bits we have qubits, which use superposition and entanglement. A superposition is a state in which a qubit, that is a quantum, is simultaneously present in its every state. In other words, we don’t have a distinct state of zero or one but all states at the same time. This allows us to speed up the calculations. Entanglement, in turn, consists of one qubit affecting another. If we have two or more qubits, they will enter a state of quantum entanglement. We can entangle many of them, and then they are, first, in every possible state, and second, affecting one another. That way we can do calculations at multiple levels, and the power is utterly awe-inspiring. That’s the logic used for quantum computing. Today, quantum computers are very early in their development, but the quantum logic and quantum algorithms are already in use. And at this stage, that’s enough for us to accelerate some calculations.

Let’s come back to the opportunities.

They exist in several areas, including molecular simulations, all sorts of optimalisations, but also artificial intelligence issues and image recognition problems. I believe we will need to be extremely explicit in whatever we are trying to achieve – since the operation of quantum algorithms is very narrow. For example, if we wanted to simulate a person, then it would be very complicated, but if focused on human immune system instead, then we would probably be able to model this system, and with this, e.g., test new medication and assess how particular systems in the human organism behave after a specific dose. What we can’t pass up are thus the simulations related to medicine, drug creation, human organism functioning. On the other hand, there are opportunities in logistics – assessment how to best reach a particular place with many variables in consideration, including jams, switching delivery order, or mixing delivery and collection. In the financial domain, new possibilities are definitely going to appear, e.g. in calculations related to buying and selling securities or financial assets in a broad sense. And let’s not forget digital assets either, such as Bitcoin or NFT.

In what way will quantum technologies influence particular countries’ defence capacities?

The most popular cryptographic algorithm used today is RSA, in its 1024- or 2048-bit version. To hack such encryption, you need a very long time. It is said, for example, that some computers would need 300 billion years to brute-force this encryption. Nonetheless, we can show that with the use of certain properties of quantum-computing algorithms we are able to do it much, much faster. If we can crack the algorithm, it means that we are able to decode any coded message. Today we are already watching criminal gangs sponsored by sundry armies of the world which simply steal encrypted data. They aren’t able to decipher them today yet, but one day…

When?

In two, three years – they will be. The same goes for safety measure cracking. What needs to be added is also the ability for situation modelling contingent upon constantly updated variables. This will result in shorter response times from the relevant governmental or global institutions. That is why the impact of quantum algorithms on defence and similar fields is going to be enormous.

What about warfare? As I see it, if data gets stolen and deciphered using quantum algorithms, it may clearly influence the course of a war.

Naturally. First of all, we have the communication issue – interception and deciphering of orders. All encrypted data on troop positions, armaments, etc., may likewise be affected. Second of all, it’s worth remembering that modern-day army equipment is smart, which makes it very… non-smart. It’s peppered with computers chips, therefore reachable. On a related note, it’s exploitable – some modules and weapons can be overridden and used for other purposes. As a result, the consequences can be really far-reaching.

What stage are we at as far as the quantum computer is concerned?

Quantum algorithms are still hiding a lot of their secrets from us. I think the ongoing non-commercial work in laboratories involves quantum CPUs equivalent to a thousand qubits or more. It is usually the case that the military gets a hold of certain things faster than anyone else. Chinese scientists have published a paper where they made it known that with the use of a small number of qubits they were already able to crack RSA, the ordinary encryption algorithm. The question is how many qubits they in fact used. As they wrote, they were able to crack 48-bit RSA with 10 qubits. Claiming they could also crack the 2048-bit version. Unconfirmed as it is, if the possibility to do so existed, we would already be witnessing a repeat of WW2 Enigma and its cipher.

When will the breakthrough come?

In my opinion, we are two or three years away from building a quantum computer. We moved on from encryption-cracking theories to practice. The question is how quickly the most popular encryption will be cracked. This subject is separate from whether we’ll have a genuine quantum computer, since we don’t need a quantum computer to be using the tools of quantum logic. The game is not about who builds the first actual quantum computer. I would be more concerned about the quantum algorithms that haven’t yet seen the light of day but can markedly speed up calculations.

What is this race all about then? And what will the entity that uses the quantum-logic abilities first gain?

Among the bad things, they will be able to decipher… everything. Quite frankly, it would be more or less a disaster. In other things, the issue of a calculation speed-up and process modelling is at play. Currently, the gain would probably be much larger when it comes to the challenge-related than the opportunity-related side of things.

The issues related to quantum computing and to a possible quantum-logic-tool-use revolution are going to feature in the Krynica Forum 2023 agenda. Piotr Ciepiela, EY Partner, a global leader in security architecture and emerging technologies, is going to be a Krynica Forum 2023 guest.