“Google has shown a chip for a quantum computer, which is trillions of times faster than a normal one. Can he change the world?”, — write: epravda.com.ua
At the beginning of December 2024, the Google Quantum AI division presented a new chip for quantum computing – Willow. During the presentation, special emphasis was placed on its high performance: the processor is faster than the most powerful modern supercomputer and is able to solve a key problem of the industry that has remained unsolved for the past 30 years – quantum error correction. Moreover, Willow partially confirms the multiverse interpretation of quantum mechanics. The many-universes interpretation of quantum mechanics suggests that each quantum event creates parallel universes. For example, if a particle has several possible paths, the universe “branches”, and at each branch the particle chooses one of them. All these parallel universes exist at the same time, but we only perceive the one we are in. However, among representatives of the industry, the appearance of the new processor did not cause the hype that the media created. Scientists and specialists in the field of quantum computing met Willow without much enthusiasm, seeing it as another, albeit important, step in the development of quantum technologies. For them, Willow is a good processor, but not the revolutionary breakthrough that the media called it. Exaggerating the significance of new products is common practice in the tech world. And it doesn’t matter if it’s the presentation of a new iPhone or a chip for quantum computing – they’re all “the best ever”. Google was no exception, presenting another achievement as the main discovery of the decade.Advertisement: However, despite the pathos in the presentation, Willow’s merits cannot be devalued. Google’s new chip has drawn attention to a booming industry and is another step toward a world where computing based on the principles of quantum mechanics could become commonplace. What Google actually showed in its Willow and how quantum computers can change the world, for which a real race between states has already begun. How quantum computers work and why Willow did not become a sensation Quantum computers work according to the principles of quantum mechanics, which is fundamentally distinguishes them from ordinary computers or smartphones. The main difference is that instead of traditional bits, which can only have a value of 0 or 1, quantum computers use qubits (quantum bits).Advertisement: Due to the phenomenon of superposition, qubits can simultaneously be in states 0, 1 or their combination. This allows quantum computers to process a huge number of options at the same time, which gives them more computing power compared to conventional computers. Moreover, the real advantage of a quantum computer is growing exponentially. For example, the performance of eight qubits of a quantum computer exceeds the performance of a conventional computer by 32,768 times. “Quantum computing is, to some extent, a new stage of technological evolution. For the first time, we are moving from binary logic to quantum logic, which brings us closer to the principles of computing that occur in nature itself,” explains Mykola Maksymenko, co-founder of Haiqu, a software startup for quantum computers. Willow chip developed by Google Quantum AI Google In recent decades, quantum computing has gone from a theoretical concept to a full-fledged industry with billions of investments. Key players in this market include Google, IBM, Microsoft, Amazon and Intel. Even in this complex field, which is still little known to the general public, there is fierce competition, which forces companies to regularly demonstrate significant progress. Google with its Willow quantum chip was no exception. Yes, the real sensation of the last week was the statement about Willow’s computing capabilities. According to Google, this chip can solve a problem in five minutes that would take the fastest supercomputers 10 septillion (a number with 21 zeros) years. The founder of the Google Quantum AI team, Hartmut Neven, even suggested that such indicators could indicate that calculations are taking place “in many parallel universes at the same time.” Read also: From rise to fall – one chip. What’s going on with Intel? However, as you know, the devil is in the details. Bold statements of the company caused skepticism in part of the scientific community. The question is how to measure quantum superiority. Google uses the “Random Circuit Sampling” test, in which a quantum computer performs a series of random operations and generates a distribution of random numbers. The idea is to show that a quantum computer can do this faster than a classical supercomputer. The problem is that this test is not a real algorithm in the classical sense. “Algorithms usually work with real input data, turning it into a useful result. In the case of Willow, the algorithm is just a set of random operations and the result of random binary numbers,” Maksymenko says. As a result, Google’s new quantum chip has also come under criticism from scientists regarding the multiverse theory. Some scientists, including the American theoretical astrophysicist and science writer Ethan Siegel, consider Willow’s result accidental and do not associate it with the idea of the existence of parallel universes. Interestingly, in 2019, Google already claimed quantum supremacy with the previous Sycamore processor, which completed a task (random number generation) in 200 seconds, while a classical supercomputer would have needed 10,000 years for it. However, IBM later provided evidence that the world’s most powerful supercomputer can almost keep pace with Google’s new quantum machine. Read also: Chips per trillion: how Nvidia became the main manufacturer of “iron” for artificial intelligence One of the key moments of Willow’s presentation was the demonstration of an algorithm that allows you to create one logical qubit based on many physical quantum bits. Google emphasizes that such an aggregated qubit turned out to be much more stable than each individual physical one. In fact, the realization of a logical qubit is an important step forward that confirms the possibility of overcoming the limitations of physical quantum bits and opens the way to more reliable quantum computing. However, Maksimenko believes that the Heron chip from IBM is similar to Willow in certain parameters, and in some places even surpasses it. “We can expect that IBM will also soon present an implementation of the logical qubit. Given the open access of IBM’s quantum computer to scientists, we can see this result in the coming weeks,” explains the co-founder of Haiqu. The revolution is canceled Willow’s presentation and additional publication research in the journal Nature did not greatly surprise the scientific community. It was known about the development of the Quantum AI division already in the summer of this year. But in the media space, the Google team decided to announce itself more loudly. Hype and talk surrounding a new “breakthrough” quantum processor has sent shares of Alphabet, Google’s parent company, up 25% from September lows. The day after the presentation, Alphabet’s capitalization narrowed the gap with the rest of the “Big Seven” companies (Apple, Microsoft, Amazon, Nvidia, Meta and Tesla). The heated topic of quantum computing also affected the value of shares of other projects related to quantum technologies ologies: Rigetti Computing Inc. or D-Wave Quantum Inc. However, Willow’s actual results and impact on the industry are somewhat more modest. “Those who follow progress in the field of quantum computers understand that advances in this field are the result of incremental improvements in technology, without any magic or the influence of parallel worlds. This progress is based on many years of work by experimenters and theorists at Google who were able to double number of qubits and improve their stability fivefold,” Maksymenko emphasizes. Read also: The father of chips: how the founder of TSMC brought Taiwan to the technological map of the world At the moment, it is too early to talk about the rapid and full integration of quantum computers into human life. Most of them are bulky and resemble futuristic lamps or installations from science fiction: a huge cylinder with numerous copper wires, tubes and gold plates that stretch from the ceiling to the floor. This is a cryostat that cools qubits to temperatures close to absolute zero (–273.16 degrees Celsius). Dilution Refrigerator at Amazon’s Quantum Networks Lab Getty Images In general, the quantum computing industry is undergoing a stepwise evolution. At the same time, work is carried out not only on “iron”, but also on software. For example, the startup Haiqu, in cooperation with the British bank HSBC, created an algorithm that allows you to download real data, for example, from financial markets, to a quantum computer on a scale that is already of practical use. It is about working with tens or even hundreds of qubits, which was previously impossible. The algorithm allows financial institutions, such as banks, to take the first step towards calculations on quantum computers, in particular Monte Carlo simulations, which require large resources on conventional computers, Maksymenko explains. Despite the contributions of major technology companies and startups to quantum computing, progress in the industry will be gradual. Companies will step by step increase the number of qubits, work with the stability of calculations, while simultaneously trying to integrate the technology into practical areas of application. According to surveys by McKinsey consulting company, 72% of technology company executives, investors and scientists in this field believe that a fully fault-tolerant quantum computer will be created by 2035. The remaining 28% predict that this milestone will become a reality no earlier than 2040. Be that as it may, quantum computers are already becoming a strategic industry for most economically developed countries. Quantum computing as part of national security Quantum computers will not be able to completely replace classical computers, since they have no advantages for conventional computing. However, they are much more effective for tasks that require parallel calculations to obtain approximate results, which is almost impossible to implement on classical devices. As the Haiqu co-founder points out, quantum computers are particularly useful for molecular simulations, protein folding, aerodynamic and hydrodynamic problems, as well as optimization, logistics and defense. Separately, it is worth highlighting the potential of using quantum computers to solve problems that are the basis of modern encryption algorithms. In other words, quantum computing could make modern encryption obsolete, posing serious risks to military and economic systems. This potential threat, known as Q-day, involves the possibility for hackers to compromise critical systems such as power grids or financial transactions. Read also: From buttons for TVs to 70% of all iPhones: how Taiwan’s Foxconn conquered the world Most experts miss the possibility that Q-day will occur no earlier than the 2030s. The World Economic Forum estimates that more than 20 billion devices need to be upgraded or replaced to protect against this threat. Although quantum computing is still largely theoretical, governments are actively preparing for the potential challenges. Today, more than 20 countries have national quantum initiatives, but only a few are capable of producing quantum computers due to technology export restrictions and insufficient funding. In total, 42 billion dollars were invested in quantum technologies, which is a third of investments in artificial intelligence for the year 2022.