What is Quantum Computing and Why Should We Care?

The idea of using quantum effects to improve the speed and security of computational processes is not new. It has been around on the crossroads of Physics, Computer Science, and Mathematics for nearly four decades. From the outset, Richard Feynman’s paper, “Simulating Physics with Computers” (1982), still an inspiring, easy-to-follow reading, formulated its entire programme: resorting to quantum reality as a computational resource.

The motivation is clear: Classical computer technology is running up against fundamental size limitations, leaving a lot of hard, complex problems out of reach of all sorts of classical supercomputers one can dream of. Computing the prime factors of a number, an essential procedure underlying most popular cryptographic mechanisms, provides a good example of what counts as a hard problem: one whose resources consumption (time, memory) grows exponentially with the size of the input. Factoring a 30-digit number takes around one month in a massively parallel computer network. Attempting to do the same for a 400-digit number will take about the age of the universe. Doable, but uninteresting.

Recognising that information is a physical entity and computation a physical process, quantum computing brings together two of the last century's main intellectual achievements: Computer Science, pioneered by A. Turing in the early 1930’s, and Quantum Mechanics, which was formulated slightly before. The former, however, essentially progressed by abstracting from the physical reality. Although key to its success, such an abstracting process was carried out to an extent that its origin was almost forgotten. Quantum mechanics, on the other hand, ubiquitously governing the behaviour of all current electronic devices, had no influence on the computational model itself. Until now.

But which quantum-mechanical effects are harnessed as computational resources in a quantum computer? Basically three:

• Superposition, which allows a quantum memory to hold information of many classical states “simultaneously”, each of them carrying a specific weight, or amplitude,  which dictates the probability of  becoming manifest upon observation;
• Interference, through which different superposed states combine in a wave-like fashion, with positive and negative amplitudes cancelling or reinforcing each other;
• Entanglement, witnessing strong correlations between different parts of a system such that an action performed on one of them can have an immediate effect on the other, even at an arbitrary long distance.  

If superposition entails some form of natural parallelism, making possible the simultaneous computation of a number of solutions to a problem that exceeds the number of atoms in the visible universe, such a magic is refrained by a basic fact: when reading (or, in the physicist’s jargon, measuring) the outcome of such a process, only one of these solutions is picked up at random, all the other being discarded. There is no magic indeed, but careful engineering: quantum programming amounts to engineer this potential parallelism with suitable forms of interference to come up to the relevant outcomes with high probability.

No magic, but, nevertheless, there are remarkable achievements. For example, prime factorization, mentioned above, can be carried out in a quantum computer in polynomial time. This means that, once such computers became a reality, all cryptographic schemes based on this problem, that is most procedures underlying current secure communications, from defense to e-commerce, will collapse.

Working on multilateral regulatory frameworks, promoting scientific literacy, and enforcing effective open-science policies at all levels of the international community are essential steps towards a more inclusive future. And, certainly, the role of the United Nations in this process cannot be underestimated.

Still, this day is probably several decades ahead. But for data supposed to remain secret for extended periods, as often required, for example, of state-classified information, the threat cannot be swept under the carpet.

Indeed, research on quantum technologies is moving (very) fast. The first operational and commercially available applications are already in place. For the last ten years, the viability of quantum computing has been demonstrated in a number of cases, its utility discussed across industries, and proof-of-concept implementations proved effective in solving, in a small time frame, problems with no classical known solution. Furthermore, efforts at national or international levels to further scale up R&D are in place. And a harsh, across-industry race has begun to secure quantum talent, build quantum skills, map real-world problems to quantum algorithms, and capture disruptive advancements into questionable intellectual property rights.

In the (near) future, quantum computing will have an increased impact on several application areas. First and foremost, in the simulation of complex systems. For example, figuring out properties of specific molecules whose simulation is beyond the reach of current best supercomputers, could be addressed by a quantum device with a few hundred well-behaved qubits. The consequences for the pharmaceutical sector are easy to grasp. Secondly, in finding optimal solutions to complex modelling problems, leading, for instance, to better predictions of the behaviour of fluids (e.g., in floods) and to the design of optimized energy-efficient artefacts. Or to tackle typical search and optimization problems that both governments and companies face, e.g., in scheduling or dynamic resource allocation.

If the potential is huge, so are the challenges involved.

A quantum divide is looming at (a short) distance as companies that have started to invest heavily in quantum computing are already patenting many of the techniques, including some based on academic research. Even worse, because of the massive investment required, entrenched giant companies are less likely to be disrupted by competition from small start-ups than in the classical computing landscape. The quantum divide may multiply present inequalities and social disfuctions at the global scale: pharma research provides an excellent, terrifying, example. The quantum internet may be another, if built in a close, exclusive way, under governmental or private control (the temptation is already huge in the present-day internet...).

Working on multilateral regulatory frameworks, promoting scientific literacy, and enforcing effective open-science policies at all levels of the international community are essential steps towards a more inclusive future. And, certainly, the role of the United Nations in this process cannot be underestimated.

In broad terms, however, we need to look further away. As was the case before, at the dawn of the Industrial Revolution, the key issue is not the astonishing and marvellous potential of technology. What really matters is knowing who owns it, who controls it, and to whose detriment. The 19th-century Luddites did not destroy machines out of hatred for technology: they destroyed them because they belonged to the owners of capital in whose hands the mechanical weaver was essentially an instrument of (human) subordination. In debating the social and cultural impact of quantum computing, or of AI for that matter, this is the crucial question.

By Luís Soares Barbosa, 
Deputy Director at UNU-EGOV