INTERVIEW WITH TOMMASO CALARCO
“In Terms of Understandability, It’s Similar to a Standard Computer”

Tommaso Calarco heads the Quantum Control Institute at Forschungszentrum Jülich. The physicist specialises in quantum process optimisation. | Photo (detail): © Forschungszentrum Jülich/Sascha Kreklau Mr Calarco, you campaigned for a European quantum technology research programme in 2016. Following this, the European Union invested billions in a flagship programme that was launched in 2018. Why is quantum research so important, and what social progress can we hope to see as a result?

Quantum technology can be used in a variety of applications. One huge advantage is its security: communication becomes absolutely tap-proof, for instance. The second advantage is the tremendous computing power we expect to achieve with quantum computing. In the future this would enable us to tackle problems that cannot so far be solved. For instance in sensor technology it would soon be possible to perform very precise measurements, and medical diagnostics applications could become much more accurate. An example: since quantum technology can measure the activity of neurones in real time, we could use it in the future to carry out non-invasive brain studies. Then we’d be able to understand and investigate neurological diseases better. Or let’s consider the mobility industry: atomic clocks could make navigation devices so precise that not only would they identify the street, but also the vehicle’s lane and distance from the kerb. That would increase the safety of driverless cars considerably.

Which organisations are currently working on the development and research of quantum technology?

Until a few years ago it was primarily the academic sector, in other words universities and extramural research institutions. But over the past three or four years, industry has also become involved in research and development in Europe: quite a few start-ups selling quantum computers and other devices are involved in the first phase of the EU flagship programme. And major corporations like Bosch, Bayer, BASF and the automotive industry are researching new products.

Quantum research is expensive and needs specific expertise. Is that a barrier for companies or research institutions that want to become involved?

The initial barriers for accessing quantum technology are not actually that high, in particular where the costs are concerned. Of course the European Chips Act, which is supposed to increase competitiveness in the semiconductor technology industry, is being discussed in the European Parliament at the moment. I presented a prototype in Brussels for a quantum processor– a quantum chip that was produced in Germany as a cooperative project involving several research institutions. The equipment needed for this cost several million euros. That’s far cheaper than semiconductor chips, for which the cost of production equipment runs into billions. So the technology is quite accessible in the privileged world such as Europe and North America. Admittedly that isn’t the case for underprivileged or disadvantaged countries – not because of specific hurdles, but because of general technology gaps. And that’s an ethical risk. A physicist at Forschungszentrum Jülich works on a cryostat, which cools the chip of a quantum computer. | Photo (detail): © picture alliance / Rupert Oberhäuser In what sense?

For instance quantum simulators can significantly improve the development of new materials and chemicals. But if research in these fields is only possible in privileged parts of the world, then the production and innovation capacities will only be developed there. That would be another advantage for these societies and would widen the inequality between the privileged and disadvantaged world. However these questions on the democratisation of technology do not specifically impact the quantum research sector, it’s a fundamental problem.

What could we do to achieve more openness of technology in quantum research?

The Geneva Science and Diplomacy Anticipator foundation recently announced the establishment of an Open Quantum Institute, in which we are also involved in the capacity of Forschungszentrum Jülich. The goal of the initiative is to facilitate quantum research in underprivileged countries – for ethical rather than commercial reasons, entirely as a non-profit organisation. Researchers from countries as yet unable to build their own quantum computers would have access to our machines, allowing them to work on their own problems and gain insights.

What form do you envisage for this access – will the researchers be invited to Europe?

The method is still being discussed, but one possibility for instance would be cloud-based access. The Jülich-based HPCQS (High Performance Computing – Quantum Simulator) programme aims to provide external access to a quantum simulator that’s hooked up to the Jülich supercomputer as of next year. Once this infrastructure is up and running, it could potentially be used by researchers from other countries as well.

Let’s take this opportunity to switch from development to application: many computer and internet users already barely understand how the algorithms behind their apps work – never mind Artificial Intelligence (AI). Will this phenomenon become even more prevalent with quantum computing?

It’s true that the laws of quantum mechanics are even less intuitive. That’s Deep Tech – understanding the technology behind it requires specialist knowledge. Nevertheless the way quantum computers work is similar to standard machines, because it’s also based on algorithms that perform defined steps. So the methods they use to achieve results will be just as understandable as they are with current computer technology. The problem specific to AI is that it often delivers very accurate results – for example in some cases AI analysis of x-ray images provides far more accurate cancer diagnoses than human specialists. But at the same time it isn’t clear where and exactly how it arrives at this result. Things won’t be any different with quantum AI either – and the time for quantum machine learning will come, even though it might still take at least ten to 15 years for the technology to become viable.

Meanwhile lots of ethical debates have emerged in relating to Artificial Intelligence. For example, unbalanced datasets result in programs that reproduce discrimination. What ethical conflicts do we need to prepare ourselves for when quantum technologies are ready to go live?

The same debates we’re already having will re-emerge in the context of quantum computing, because the technology increases computing power immensely and can therefore speed up Artificial Intelligence as well. Apart from that, there is of course the fundamental risk that new developments could be used for malicious purposes, for example arms development. That’s a hot topic amongst scientists: how can we ensure that our own discoveries are used for benevolent purposes? But in both instances the ethical questions are not new, they have been discussed in connection with technology development for a long time. Until a few years ago, lots of scientists were convinced that they were neutral. But that doesn’t release us from the responsibility of counteracting abuse.

Are there efforts at EU level to meet this responsibility?

For instance in the “Coordination Actions” of the flagship programme there are workgroups that focus on responsible research and innovation, and are in constant communication with the quantum community. An early result of this work is the already-mentioned Open Quantum Institute, in which many European institutions are participating.