Have you ever heard of qubit, computers and quantum computers? If so, you are on time. If not, you will have to get serious about it because their commercial deployment is underway.
Some see it as the end of the blockchain; others see it as a golden goose or simply a way to exceed the limits of conventional computing. Everything finally seems to indicate that Google would be on the way to quantum supremacy.
The qubit or quantum bit
This is what the qubit causes in its path. IBM, Intel, Microsoft, all dream at least of taming it and advancing on this technology. But what exactly is the qubit? What does it consist of? What are its advantages and disadvantages? And what are its commercial opportunities today? Forecasting AI has therefore taken up the subject and provides you with some initial answers.
First, the qubit, for quantum bit, is a quantum state that represents the quantum information storage unit. Used as a value in quantum computing, which aims to use the laws of quantum physics such as entanglement or superposition to perform faster calculations, the qubit is different from the bit, a unit of binary information (0 or 1) used in computing as we know it more precisely today.
The qubit is more powerful than the bit
A bit can therefore only take one state, and only one, at a time, to choose between two values such as "0" and "1". The qubit differs from the bit in that it consists of a superposition of two basic states such as "0" and "1". To sum up, the power of qubits is exponential and can make it possible to process an operation much faster than a binary computer system.
And for good reason, it would require four operations with a conventional computer versus only one with a quantum computer for an operation with two bits. The quantum computer will process all operations at the same time thanks to the superposition of states (2²). A binary computer will have to decompose all operations one by one since the bit can only be one state at a time (so either 00, or 01, or 10, or 11 for an operation with two bits). Four simultaneous operations against one, it goes without saying that you already know the winner in terms of execution speed.
Powerful but unstable, the qubit...
Of course, I see you coming... What are we waiting for to move from binary to quantum computing? The qubit has a natural exponential power, but it also has significant limitations.
Taking advantage of the laws of quantum physics, the qubit has not been able to overcome some of the pitfalls that many scientists are currently working on. Qubits remain unstable elements that are extremely sensitive to environmental variations. This phenomenon is called decoherence. It simply stops the state of superposition of qubits and thus puts an early end to the calculations undertaken.
Qubit vs cryptography
A simple and minute variation in temperature or light, magnetic interference can precipitate its "fall". The size of the machines or the theorem of impossibility of quantum cloning are other disadvantages to be added to decoherence even if the computing power and confidentiality of the exchanges increase (a qubit can neither be read nor duplicated without fixing its shape).
And then the advent of quantum computing is likely to call into question even the discipline of cryptography. The blockchain uses, for example, RSA and AES keys that can be broken by the Schor and Grover algorithm. Impossible because it takes too long to do with a binary supercomputer, but this task could be entrusted to a quantum computer that would break the systems. The security system we know today would be upset by this, but could also take advantage of quantum computers to generate keys that are ever larger and longer to decipher.
Until quantum computing comes into the home in the same way as conventional computing does now, and it will not be until we succeed in stabilizing the coherence of a qubit over time, some countries and companies are relying heavily on this technology, even aiming for quantum supremacy. But what exactly is quantum supremacy?
This expression could be summarized as follows: to prove that a quantum processor would solve a series of computer calculations that even the most powerful supercomputers in the world would be unable to perform in 10,000 years! Would Google have succeeded in this crazy and insane bet? On Friday, September 20, the Financial Times reported this information after a Google researchers' publication on NASA's website was published and then removed in the process. And since then, many copies of this article have been circulated on the Internet and the media have rushed to this information to make it their own.
Sycamore, Google's quantum calculator
To achieve such a feat, according to the information read here and there, Google would have used a quantum computer called Sycamore. The beast would have more than 50 qubits (54 to be exact, 53 of which were actually used). His specialty (because quantum computers like to have a specialty...)? Simulation of the behaviour of a random arrangement of quantum circuits.
But it doesn't matter! Because such information is beautiful and always to be confirmed or denied and neither Google nor NASA, which would have broken an embargo out of clumsiness at best, have so far wished to respond to the requests of the journalists who are coming.
The announcement of Google's quantum supremacy would still be a significant step forward for the Mountain View firm and would give it a clear advantage over its competitors by offering the fastest machines on the market in terms of calculations. Able to calculate the incalculable! There is no doubt that IBM could also achieve this in the near future, but being the leader in a sector as advanced, technological and high value-added as quantum computing offers you credibility, legitimacy and notoriety with your prospects and customers.
However, this advantage, which is an advantage when it comes to offering for sale a computing power never seen before, is not yet decisive. Indeed, the quantum decoherence and ultra-sensitivity of qubits to their environment make it almost impossible to operate computers with a large number of embedded qubits or, obviously, to sell quantum computers to a general public.
IBM, a quantum computer available on the cloud...
Excluding homes and commercialization to the general public for some time to come, quantum computing is now entering the cloud. While Google wants to reach and focus on quantum supremacy, IBM has probably wanted to be more pragmatic by providing a 53 qubits powerful quantum computer on the cloud. Soon to be in service, this computer could solve new problems and enchant the best forecasters thanks to its computing power.
One is trying to achieve quantum supremacy to impress its world, the other to beat its competitors to propose breakthrough innovations to its customers. Of course, behind these actions, the two American giants are engaged in a real trade war.
And IBM may have taken a step ahead of Google by already offering its computer in the cloud. Financially inaccessible, even for the largest companies, because it is difficult to install and operate, and therefore rare, the quantum computer will not flood the market at first. So, offering such a service for rent to its customers is probably the best possible option. This is what IBM finally did before Google.
But who would like that? What would be the commercial opportunities for a quantum supercomputer? You may not think about it, but of course Google and IBM already have their ideas!
What applications for quantum computing?
Such computing power will of course not leave forecasters on all sides unaffected. A natural exponential power. The pharmaceutical industry would not be against using it in its research. Finally, other potential customers include public markets, administrations and other cities that have to manage road, public transport or communication networks, for example, insurance companies, banks and financial markets that depend on forecasts and fear for their digital security. Not to mention the supermarkets, of course, which depend on sales forecasts or stock forecasts.
In other words, there are many, if not juicy, opportunities for those who will succeed in doing well in the quantum computing market. Google and IBM understood it well and were the first to draw with their powerful resources. But, without a doubt, others will also want their share of the cake and will invite themselves into this crazy race.
blog comments powered by Disqus