Vincenzo Savona, head of the Laboratory of Theoretical Physics of Nanosystems. 2025 EPFL/Murielle Gerber- CC-BY-SA 4.0
Vincenzo Savona, head of the Laboratory of Theoretical Physics of Nanosystems, discovered a passion for quantum physics while learning about semiconductors. He took some time out to share his insights into the field, which he's been teaching at EPFL for over 20 years.
How do you combine quantum and classical physics in your teaching?
I like to remind students that classical physics is a branch of quantum physics, and not the other way around. Quantum theories were first introduced exactly 100 years ago, and today we know they can explain everything that goes on around us. The equations that describe our environment were written in 1925 and they're still perfectly valid today.
So why do we still teach classical physics?
Mainly for practical reasons. The theories of classical physics accurately describe many processes and are used to perform engineering calculations, develop computer models and explain a broad range of phenomena. Classical equations are easier to work with than quantum ones, making them suitable for numerous applications. But already back in 1900, Max Planck hypothesized that quantum physics could explain things where Newton's and Maxwell's theories fall short. Then, for the following 25 years, other physicists - including Albert Einstein, Niels Bohr, Werner Heisenberg, Wolfgang Pauli and Erwin Schrödinger - explored this idea and developed equations that are now considered the building blocks of theoretical quantum physics.
So that's the background. But one century later, quantum theorists like you still have a lot of work to do!
Absolutely. And that's exactly what makes quantum physics so amazing - its theories can explain everything that consists of light or matter. They give us a larger toolbox than the one provided by classical physics, even though we haven't yet mastered all of those tools. There are some processes that we know can be explained by quantum physics, but we haven't yet found the right equations.
Like what, for example?
One of the biggest challenges for physicists today is to discover a single unifying theory that covers quantum physics and Einstein's general theory of relativity, which describes the process of gravity. We've made a lot of progress, but we aren't there yet. Another example much closer to home is high-temperature superconductivity. We still don't fully understand the underlying mechanisms of this property of materials, but it's used in a lot of applications, such as magnetic resonance imaging.
So quantum physics is about more than just tiny particles?
People tend to think that quantum physics relates only to atoms, electrons and photons, but that's not true. For example, we know that the stability of matter owes to the laws of quantum physics. Under classical physics, electrons that rotate around a nucleus would lose energy and collapse into the atom in a nanosecond, causing the matter as a whole to implode. Only quantum physics can explain why that doesn't happen. On a larger scale, astrophysicists can use instabilities in the cosmic microwave background to observe signs of quantum phenomena that originated with the Big Bang.
Looking beyond the pure theory, what are the research applications of quantum physics?
There are a lot. At EPFL, we're developing quantum-related applications mainly in computing, communication systems, modeling and sensing.
Are many students drawn to the field despite its complexity?
Yes. I must say that my master's class in quantum computing is very popular. We also get many requests for master's projects in the field. Fortunately, we have contacts with numerous external partners, including companies, through EPFL's Center for Quantum Science and Engineering (QSE Center), and that creates opportunities for those students.
Does EPFL stand out in this field?
Yes, without a doubt. It's been a priority area for our School for more than a decade, and we've hired close to a dozen new professors. The opening of the QSE Center in 2021 was a major step forward, and the following year we introduced a master's program in quantum science and engineering, which immediately received many applications. I'd also point to the Advanced Science Building that will soon be built. It will provide our researchers with extremely powerful equipment, giving us a formidable resource for moving this field forward. As a result, EPFL - and Switzerland in general - are very well-positioned internationally in the quantum applications I mentioned earlier. Our School is also focusing heavily on the development of quantum algorithms, which will have uses in numerous areas.
Leveraging the full potential of quantum computing is just as ambitious as the Moon landing was.
So the plan is to develop the algorithms before the computers are ready - isn't that backwards?
I'd compare the research we're doing in quantum calculations and computing to the pioneering work done in electronics in the 1940s - we see now how that evolved in the subsequent decades. Progress on quantum computers is being made at a dizzying pace. Engineers are exploring several approaches in parallel and creating numerous prototypes. We don't yet know which technology will prevail. But there are clear hurdles we'll need to overcome, such as correcting quantum errors and developing special software for testing quantum computers.
What will quantum computers be used for?
The idea isn't just to build machines that are faster than the ones we already have - what we're looking at is a paradigm shift in how data are processed. Many potentially useful calculations can't be done with classical computer architecture because the amount of time and resources they require grows exponentially with the size of the operation to compute. Quantum computers could perform some operations much more efficiently. For instance, they'd be perfect for creating digital models of matter, or for running optimization calculations, since they can process large volumes of data simultaneously. Some physicists, including Zoë Holmes at EPFL, are exploring quantum machine learning, which could dramatically improve the performance of artificial intelligence.
Are AI and quantum computing competing approaches?
They're being developed at the same time and feed into each other, which is paving the way to incredible advancements. Engineers are using AI to improve the design of quantum computers, for example.
Speaking of design, why do these computers look so strange?
Quantum chips must be cooled to temperatures close to absolute zero (-273.15 °C) in special cooling units for them to operate. Data are transmitted into and out of the cooling units using microwaves, which are carried in the tubes that give these computers their hallmark shape. My colleague Tobias Kippenberg is working to develop transducers that can operate inside the cooling units, thus enabling fiber optic cables to be run right into the system core. If he's successful, that would be a huge technological leap forward.
When do you think we'll start to see the first large-scale applications?
It wouldn't be realistic to make a concrete forecast at this point. This is a planet-wide, highly collaborative effort - along the lines of the Moon landing! We know that quantum technology has amazing potential, so we have a moral duty to try to realize that potential. It'll involve a lot of trial and error, a lot of false starts, but we'll learn a great deal from those mistakes. I see it as a sort of Pascal's wager - we have nothing to lose by believing, so we might as well do everything we can to make it happen.
Cryptography: "No, we are not doomed"
Does quantum threaten encryption?
Every few months, a new wave of headlines warns us that quantum computing is about to destroy digital security as we know it, that within three, or five, or perhaps ten years, quantum machines will supposedly be powerful enough to break the cryptographic backbone of the internet. That means everything from your WhatsApp chats to your credit card transactions could be made public.
At the heart of this concern lies RSA public-key cryptography, a widely used system whose security hinges on the difficulty of factoring large integers, a task considered unfeasible for classical computers. But Shor's algorithm, a brilliant quantum algorithm that, in principle, can factor such numbers exponentially faster than any classical method, could make RSA effectively useless.
So, should we panic?
I don't think so. First, the cryptography community has been preparing for this for years. Researchers have developed an entire new generation of so-called post-quantum cryptographic algorithms, based on mathematical problems that even quantum computers can't crack. Some of these algorithms have already been endorsed as standards by the US National Institute of Standards and Technology (NIST). If the moment of reckoning ever arrives, swapping out RSA for a post-quantum alternative will, in most cases, require nothing more than a software update.
Second, the kind of quantum computer needed to break RSA is still the stuff of theory. Estimates suggest it would take around one million high-quality qubits all operating in perfect harmony. Today's quantum devices are nowhere near that scale, and achieving such uniformity and control remains one of the biggest technical challenges in the field.
And even if such a machine did emerge, there's a final twist: solving a hard factoring problem is not the same as solving the RSA problem. RSA numbers are specifically crafted to be among the most challenging to crack, even for quantum algorithms.
So, no, we are not doomed. Not yet, and likely not anytime soon. The cryptographic world is evolving, and quantum computing, for all its promise, is not an unstoppable force-at least not today. For now, I sleep well. And so should you.
