We are experiencing the second quantum revolution. The first revolution coincided with the discovery and understanding of the laws of quantum mechanics in the early 20th century, an example being the development of lasers. The second quantum revolution took place in the 1980s, with the discoveries of physicist Richard Feynman and Alain Aspect’s quantum correlation experiments, which led in the early 2000s to the first gate between two superconducting qubits (the unit of computation in quantum computers) and later to the development of trapped ion technology, with which even 20 or 30 qubits could be held together.
Today, among the companies engaged in the development of quantum computing, the largest focus is on the development of superconducting materials technology, and not on trapped ion technology. These include Intel, IBM and Google.
Unlike a classical bit, which can only have the value 1 and 0, the qubit can take on several values simultaneously thanks to a phenomenon called quantum superposition. In the case of Google’s chip, this is done by cooling a superconducting metal to temperatures close to absolute zero. In this way, each qubit added to the system exponentially increases the amount of information that can be stored in the system, which can be exploited to greatly speed up computational problems that are difficult to solve with binary computers.
Google and Willow
The first to have launched a quantum computing prototype was IBM in 2019, with its Ibm Q System One. Today, however, it is the turn of Google, which, from its Quantum Artificial Intelligence Lab, focused on the study and research between AI and quantum computing, has launched a new chip, Willow.
Back in March 2018, Google’s research in its Quantum AI Lab had made it possible to produce a 72-qubit quantum computer chip, called Bristlecone, capable of performing in 3 minutes a processing that at Ibm Summit (the world’s fastest non-quantum commercial supercomputer) would have taken 10,000 years. The experiment would have been performed with a 53 qubit processor codenamed Sycamore. A few months later, Google had confirmed the achievement in a study that appeared in Nature, explaining that it had performed an extremely complex calculation in 3 minutes and 20 seconds using its 53-qubit Sycamore quantum chip.
Earlier this week, in a Google presentation post, the company stated that ‘Willow performed a standard benchmark calculation in less than five minutes, for which one of today’s fastest supercomputers would take 10 septillion, a number that far exceeds the age of the Universe’.
In quantum computing, errors represent one of the biggest challenges: qubits, the units of computation in quantum computers, have a tendency to rapidly exchange information with their environment, making it difficult to protect the information needed to complete a calculation. Therefore, generally, the more qubits used, the more errors could occur. But, according to Google’s announcement, ‘we have published results in Nature showing that the more qubits we use in Willow, the more we reduce errors and the more quantum the system becomes‘.
With 105 qubits, Willow would now have best-in-class performance: quantum error correction and random circuit sampling. Willow would thus represent an improvement of about five times over Google’s previous chip generation.
To measure Willow’s quantum performance, Google used a standard benchmark, Random Circuit Sampling (RCS), created by the Google Quantum team. The researchers were thus able to demonstrate that the quantum computer under test is indeed capable of performing calculations that a traditional supercomputer would not be able to perform, succeeding in exploiting the ‘parallel’ calculation capabilities of quantum computers to the full and proving that it has a real advantage over a traditional system.
The quantum computing market
Although it will still be some time before a fully-functioning quantum computer becomes a reality, investors are showing a growing interest in this area of technology that promises to one day enable calculations at incredible speeds and enhance the creation of artificial intelligence models. Even Europe, with its Quantum flagship initiative, has earmarked a billion euros over ten years starting in 2018.
According to the Milan Polytechnic’s Quantum Computing & Communication Observatory, from 2019 to date, a total of USD 5.9 billion has been raised, 56% of which is concentrated in America compared to 29% in Europe and 10% in Asia. In Italy, the funds raised by start-ups active on Quantum Technologies reach €12 million over the 2023-2024 horizon compared, for example, to the €255 million raised in France over the same period. Despite this, the first Italian investment fund on quantum technologies was set up in 2024, precisely to narrow the gap with other countries.
Quantum computing and IA
Artificial intelligence and quantum computing are not disconnected topics, far from it. The following are the words of the author of Google’s post on Willow’s presentation, Hartmut Neven, founder and head of GoogleQuantum AI: ‘Sometimes my colleagues ask me why I left the burgeoning field of artificial intelligence to focus on quantum computing. My answer is that both will prove to be the most revolutionary technologies of our time, but advanced artificial intelligence will benefit greatly from access to quantum computing. That is why I named our lab Quantum AI. Quantum algorithms have fundamental scalability laws on their side, as we are seeing with RCS. There are similar scalability advantages for many fundamental computational tasks that are essential for artificial intelligence. So quantum computing will be indispensable for collecting training data inaccessible to classical machines, for training and optimising certain learning architectures, and for modelling systems where quantum effects are important. This includes helping us discover new medicines, designing more efficient batteries for electric cars and accelerating progress in fusion and new energy alternatives. Many of these future revolutionary applications will not be feasible on classical computers; they are waiting to be unlocked with quantum computing. (image from Google)
ALL RIGHTS RESERVED ©