Benefits and challenges for unleashing potential of quantum technologies

Event report

The session announcement said that the EU plans to develop the quantum-safe internet all over Europe by 2030, and asked what the potential benefits and challenges for society could be. Mr Rosario Fazio (Head of the Condensed Matter and Statistical Physics Section, Abdus Salam International Centre for Theoretical Physics in Trieste), session moderator, reminded that quantum computing and communications do not use bits as basic units of information (with two possible states: 0 or 1), but quantum bits, or qubits.

Mr Dimitris Angelakis (Principal Investigator, Centre for Quantum Technologies in Singapore and Assoc. Professor, TU Crete Greece) explained that, as quantum computing relays on manipulating information at the sub-atomic level, rules of quantum mechanics allow the qubit to be in 0 or 1, but also in both states at the same time (which is known as superposition). Since conventional computing is reaching its limits due to the end of Moore’s law (it is increasingly difficult to further scale down the size of chips, to fit more of them on a silicon wafer and increase computing power), quantum computing will allow next levels of computation.

Why do we actually need more computing power?

As Angelakis explained, advanced research in various areas of science and technology – from climate change to new materials and drugs – requires the ability of simulating highly complex models, which cannot be done with classical computing. In practice, the controllable quantum systems could, among other:

• Optimise operations in logistics, banking, and financial sector, transport and shipping;
• Simulate complex chemical and biological processes such as proteins or new drugs, to advance healthcare;
• Simulate the design and characteristics of new materials, such as superconducting materials, and materials with special properties;
• Advance machine learning and artificial intelligence towards predictions of trends based on large datasets.

Ms Sabrina Maniscalco (Professor, University of Helsinki) emphasised the possible scientific and tech breakthroughs, due to the ability to manipulate individual atoms in systems. This could be of great value for simulations and engineering in quantum biology, chemistry, thermodynamics, communications, gravity, materials, etc. Of particular importance to the EU plans is the ability to create complex quantum networks, or a ‘quantum internet’.

Reflecting more on machine learning, Mr Gioacchino Massimo Palma (Professor, University of Palermo) explained that it can help improve quantum computing. One of the main problems with achieving a controllable quantum machine is the undesired interaction between the qubits and their environment (i.e. the entanglement of particles in the system and outside of it). Classical machine learning can describe and classify interactions of particles within the system and with the environment. This can contribute to devising the ‘open quantum systems’ in which the quantum system is allowed to interact with its environment. On the other hand, quantum computing can inspire and advance machine learning, in forms such as hybrid quantum-classical machine learning, or quantum reservoir computing.

The gap between the reality and hype

Responding to a question from participants, Maniscalco clarified that the current level of applicability is still modest. According to Angelakis, current quantum systems can perform certain useful mathematical tasks, such as sampling problems which classical computers would need thousands of years to do, but there are still no public applications. It is expected that the breakthrough in quantum hardware (and, in particular, in optics), new approaches to quantum computing (e.g. photonics, superconducting qubits, trapped ions), and the ability to manipulate particles ‘one-by-one’, will bring us closer to controllable quantum computers.

Angelakis stated that the estimated global investment in quantum in 2021 is US$22.5 billion, and informed about the global rush for patents.

Palma and Angelakis agreed, however, that the general hype around quantum computing is leading us to a ‘quantum bubble’ and, in order to avoid the bubble to burst, it is of the essence to identify the real areas where quantum computing (for instance machine learning) can outperform the classical ones.

Societal risks of quantum technologies

Reflecting on a question from participants about the societal risks of quantum technologies, Palma mentioned encryption and privacy: code-breaking with quantum computers will make most of the previous encryption mechanisms obsolete at some point. As an example, Angelakis explained that a recent standard of 1024bit RSA encryption would take 10 hours to be broken by a quantum machine of about 4000 fully controlled qubits; yet, he reminded that current quantum machines have only a dozen of functional qubits, and that it may take years until we reach such computing level.

Palma also stressed the importance of equal access to quantum technologies for all of the world, and added that quantum computing will distort the job market similarly to how AI does. Angelakis suggested that the society should think about how to provide ‘democratised access’ to quantum computers, and how to prevent certain groups and countries from using quantum technologies in a new arms race.