Why are quantum computers so interesting

What does Google's quantum computer do?

There have been rumors for weeks. Now it is official. Google announced quantum supremacy on October 23, called "Quantum Supremacy". This term describes the moment at which a quantum computer is superior to conventional computers for the first time in a certain task. The evidence is considered a milestone and has now been published in the journal Nature.


Countless commands at the same time

A quantum computer should be able to perform certain arithmetic tasks much faster than the best supercomputers today. It is based on a principle that is completely different from that of an ordinary PC. It calculates with bits - with switching units that are either 0 or 1. The quantum computer, on the other hand, is based on the so-called qubit. This can be 1 and 0 at the same time and can therefore assume all values ​​in between. This enables a quantum computer to execute countless commands in parallel instead of one after the other like a normal computer - which should give it an enormous speed advantage.

The professional world has been working on prototypes for a long time. But so far these had one flaw - they were simply slower than ordinary computers. “We wanted to check whether a quantum computer really has such enormous computing power,” says Google researcher John Martinis. For this purpose, his team built a fingernail-sized special chip called "Sycamore". It has 53 functioning qubits, but has to be cooled close to absolute temperature zero at minus 273.15 degrees.

The superiority of the quanta

With the support of other teams from Jülich and Erlangen, for example, Google tinkered with it for a good year. Then Sycamore was able to solve a special task in record time: after just 200 seconds, he spat out a number that was thoroughly random. A supercomputer would need an estimated 10,000 years for this - the quantum superiority has been demonstrated. "Now we have a great future ahead of us," says Martinis happily. "Because now it is clear that the technology is more advanced than many thought."

But not all experts are convinced. Competitor IBM thinks that the problem can be solved by supercomputers in two and a half days instead of 10,000 years. In that case, the quantum superiority would still be there, but no longer as impressive as Google claims. However, the result does not yet mark the final breakthrough. Because the algorithm that Google used is only partially suitable for technical applications. For more practical algorithms, significantly more qubits would be needed than the 53 from Sycamore. In addition, experts still have to develop a sophisticated error correction for the highly sensitive qubits.

Quantum computers are not yet suitable for everyday use

But Google is optimistic. "In a few years we want to implement a chip with a thousand qubits," announced Martinis. “That would be another milestone.” The applications should initially be of interest to science, for example in the simulation of chemical reactions. Later on, quantum computers will search through databases and help with drug development.

Google hopes for advances in artificial intelligence, such as automatic image recognition. Another application, especially of interest to the secret services, is likely to be a long time coming - the decryption of digital security codes, which were previously considered unbreakable. "That would probably take 100 million qubits," says Martinis. "It will be a while before we are ready."

In any case, Germany could fall behind in terms of development. German experts are happily involved in basic research. In contrast, the economy is still holding back - and prefers to rely on the development of highly sensitive quantum sensors, for example for medicine.