Google says it has achieved an unprecedented result in quantum computing

Google has announced that it has reached "quantum supremacy", a very important and long-awaited goal in the search for alternative and more powerful systems for traditional computers. The details of the experiment were published in the scientific journal Nature and they are having discussions, because the subject is very delicate and has involved many groups of researchers and some of the largest IT companies in the world, such as Intel and IBM, which has challenged some of the statements provided by the Google working group.

Even though the term may sound high-sounding and an arrogant thread, "quantum supremacy" means the ability to solve a problem that classic computers cannot solve in a reasonable human time. Simplifying a lot: Google has asked its quantum system to perform a complex mathematical operation generating random numbers. The operation would have taken 10 thousand years to be carried out by the most powerful supercomputer available today, but Google instead got a response in just 200 seconds.

The system that made it possible to achieve this result is still at a primordial stage of development and years of research and innovation will be needed before it can be integrated into a device and used instead of traditional computers. For many researchers the result is however comparable to the first steps in a new technological era, as happened for example 62 years ago when the Soviet Union put into orbit the first artificial satellite in the history of mankind (Sputnik 1): it emitted a simple "Beep", yet paved the way for technologies that today allow us to communicate, observe the changes of our planet and study galaxies billions of light years from us.

A quantum system works differently than a normal computer, to the point that it requires a bit of mental gymnastics to understand its characteristics. Without going into too much detail and adding complications: a traditional computer – like the one on which you are reading this news (also applies to smartphones, which are pocket computers) – works by performing calculations based on "bits", information units that can take two values: 1 or 0 (think of it as a switch that can be opened or closed).

In quantum computers things are a bit more complicated, because a single object can behave like a combination of two separate objects at the same time, whether it is extremely small or is at an extremely low temperature. By exploiting this feature, researchers can construct a "qubit" ("quantum bit"), which can contain a combination of 1 and 0. Two qubits can therefore assume four values ​​at once, and so on, with numbers that grow exponentially as the qubits increase. The increase is such as to make a quantum computer enormously faster and more powerful than the computers we use today based on bits.

As you may have guessed, there must be a problem somewhere, otherwise we would already have quantum smartphones in our pockets. The problem, which has kept researchers busy for decades, is that quibits are rather difficult to manage. Keeping them together is very complicated and isolating them from the rest, to take advantage of their features, it is so difficult to have pushed researchers to use their inventiveness a lot to deal with them.

The most promising answer came about twenty years ago, when researchers in Japan developed qubits based on some superconductors. During their experiments, they noticed that some types of metals made it possible to obtain qubits that were easier to manage, if cooled to temperatures close to absolute zero (the lowest theoretically obtainable temperature, above -270 ° C).

Since then, the largest IT companies and numerous research centers often funded with hundreds of millions of dollars have adopted that method. It is for this reason that their systems are more like a refrigerator than a normal computer. They are usually large metal cylinders that reach very low temperatures inside them, where the components are to perform the calculations. It is by using a machine like this that Google has finally managed to achieve "quantum supremacy", thanks to the work of one of its research groups in the laboratories of Santa Barbara, in California.

The news of the important result was accidentally released in late September, when a version of the study was briefly published online. Its appearance had attracted the curiosity of many researchers, in a sector where academic competition and between companies is very high, and all keep an eye on.

In view of the official publication on Nature, Monday October 21 IBM had published its own article, questioning the results obtained by Google. According to the company, the processing requested from the quantum system could theoretically be performed by a traditional supercomputer in about two and a half days, and not 10 thousand years as claimed by Google. For this reason, according to IBM we could not yet talk about "quantum supremacy", because Google's system would in fact have no capacity so superior to the traditional ones. According to other researchers, the experiment would not have demonstrated much because the system was asked to produce numbers at random, a process that could hardly be used for anything practical.

Google has substantially rejected IBM's claims, conceding that in any case we are moving in a research area still little explored and that proposals to produce more realistic simulations are welcome, with a view to further advancing the research. Excluding these disputes, the Google experiment did however serve to show that a complex quantum system is sufficiently reliable to perform a task.

It will still take years before having a complete quantum computer proper, for which the uses are not even completely clear yet. Researchers believe that systems with enormous computing capabilities will enable more powerful and reliable artificial intelligence systems to be built than current ones, to create simulations in numerous fields, such as meteorology and climatology, without counting the systems to develop new drug molecules next generation.