Why does Germany need a quantum computer? Last Tuesday, in the presence of the Chancellor, an apparatus was presented in Stuttgart on which industry and research institutions will in future be able to manipulate so-called qubits - physical entities that can represent not only zeros and ones like the bits in conventional computer science, but also potentially everything in between. As we have only known since 1993, it is in principle possible to solve arithmetic problems that are too difficult for even the largest supercomputers today. Because with every additional qubit, the theoretically possible computing power doubles.
Last but not least, that inspires the imagination of those responsible in politics and research management.
Much is prompted about the blessing of this technology to secure the business location, about new active pharmaceutical ingredients or new materials for the simultaneous rescue of the climate and prosperity.
A million qubits would be good
Stay on the carpet. The quantum computer in Stuttgart can process 27 qubits - certainly a huge step forward compared to what was possible just a few years ago. But keeping qubits undisturbed is extremely difficult. For real calculations, therefore, additional physical qubits are required for error correction - and a very large number of them: In order to calculate something useful that is impossible for conventional computers, a quantum computer has to work with hundreds of thousands to several million physical qubits. For example, a panel of experts from the American National Academy of Sciences came to this conclusion in 2019 in a comprehensive report on the state of research on quantum computers. Despite all the progress, it is impossible to say whether this will ever be feasible and, if so, how many generations it will take.
Because scientific progress is not everything. The dynamism of the classic computer, which has brought us from the tube computer to the iPhone in just sixty years, was also based on a peculiar self-reinforcing interaction between scientific and economic success: New digital technology products allowed companies to earn more money, that they could then put into the development of newer technology. In more than 25 years of quantum computer research, such a dynamic has not yet started. The Stuttgart quantum computer is also purely an experimental device and will not promote the synthesis of a new anti-cancer drug any more than the current or under construction reactors of the fusion researchers will feed in electricity somewhere.
But as in nuclear fusion, the problem with quantum computing is not only that there is no money to be made with the many intermediate steps that may one day lead to success - at most, there is a gain in corporate image or national prestige. The problem is also that it does not say that the basic principle can be scaled up. A machine with 50 effective qubits (or logical qubits, as quantum computer scientists say) does not yet show that a computer with 100 logical qubits is also possible. For the latter, the said hundreds of thousands or millions of physical qubits would be necessary for error correction, and experts consider it doubtful whether this can be realized by upscaling the concepts for quantum computer hardware that are being investigated today.
However, it is also not certain that the undertaking is hopeless. Hope rests, among other things, on physical tricks that are still unknown today or new types of quantum algorithms, through which the number of necessary correction qubits may not have to be so large after all. But nobody will discover anything like this if the present, pure basic research is not carried out on the qubits and if students and young researchers are not already able to play with quantum hardware, albeit more primitive. And that's exactly why Germany needs a quantum computer.