The lab is in a secluded location, in a small, unidentified office. It has a chaotic, old-fashioned feel, a bit vintage. There are loose cables, screwdrivers, computer parts, whiteboards with mathematical formulas, disassembled computers. And inside a red bell, connected with dozens of cables, a kind of imposing half-meter chandelier that intones a soft and constant chip, chip, chip. It's hard to believe that in this secluded place, about two hours from Los Angeles, a great technology milestone occurred in October that jumped to the front page of all the newspapers. Here, at Google's quantum computer lab in Santa Barbara (USA), the company has managed to make that kind of giant lamp perform an operation to calculate random numbers in 3 minutes and 20 seconds that would take the world's most powerful computer thousands of years. That lamp houses a quantum chip and its achievement (called "quantum supremacy") is for the Leonese Sergio Boixo, chief scientist of quantum computing theory at Google, as the first flight of the Wright brothers: a still modest but opening milestone I pass to a whole new industry that will radically change the society in which we live. Rivals and independent researchers believe the comparison may be overstated, as the quantum revolution is still a long way off, but all experts agree that when it does come, it will change absolutely everything.
This is the story of a fascinating and mysterious world, where the rules of physics we know, those that govern the bounce of a ball or the fall of an apple, do not work. The subatomic world is a kind of wonderland, a tiny and strange place where Alice could be straddling seemingly incompatible situations, where what is done in one place can instantly affect an object that is far away, and where it is not. You can look with impunity because that look alters the observed object, as Andrés Cassinello, author of The Quantum Reality explains . And it's also the story of the pioneers trying to make sense of this strange world to make the smartest and most powerful computers humanity has ever known. In the race for quantum computing, countries like China and the United States and companies like IBM and Google have invested billions of dollars in building a car that, in the simile of the race, is not that it advances others, it is that he would reach the goal a few moments after the starting gun. It is what Juani Bermejo, a researcher at the University of Granada, calls “quantum power”.
The key to understanding this revolutionary technology is at the base of its operation. The computers that we know work with binary programming bits, "0" or "1". But quantum bits or cubits have three characteristics that make them special. One is the superposition, whereby two cubits can be the four combinations of "0" and "1" at the same time, which exponentially multiplies their calculation capacity. It is something similar to what happens when a coin spins: it is a combination of heads and tails, as explained by IBM research director Darío Gil. The second idea is called "entanglement" and it is very romantic: the state of interlocking cubits cannot be described independently. In Gil's example, if two intertwined coins rotate, when measuring them we will see that if one is expensive, the other will also be, and if one is a cross, the other will be the same; the probabilities are not independent. And the third idea, that of interference, is how it happens with waves in the sea, which have peaks and valleys, which can interfere with the peaks and valleys of other waves, explains Gil. This unique combination of characteristics, and so counterintuitive, makes quantum not just another category of computing we know. "It is another world, it is the first time where a fork is made in the computing category," Gil explains. "For me, we are in the most exciting moment in the world of information technology in the last 50 or 60 years," adds this Murcian expert, who is the first European to head the powerful research division of IBM in its 75 years. years of history.
There are things that classic computers do very well and that supercomputers do better. But quantum computers are in another dimension, and so we will not be able to know exactly what they will be able to do until they are fully developed. But we can intuit how they could help us. The clearest fields are those where quantum rules work outside of our reality: physics and chemistry. Quantum computers could simulate new molecules for the pharmaceutical industry that would help us, for example, to achieve drugs in record time for a global pandemic like that of covid-19. They can also improve our knowledge of how the universe originated, discover new materials, improve electric car batteries, achieve more efficient use of energy. And, while it may not seem too sexy, another relevant example is improving nitrogen fixation to produce fertilizers, which generates more than 2% of the world's CO2 emissions.
"This is the most exciting moment in the computing world of the last fifty, sixty years"
DARÍO GIL, Director of Research at IBM
The point is that, as Boixo explains, classical computing has used the same type of rules since the abacus was invented. “We have had, of course, impressive engineering and technological advances: on an abacus you have a few parts and move them with your hands, and a supercomputer has billions of parts that move billions of times per second. But the fundamental operations are the same. Now we have a new method of computing different from the methods we have been working with for 3,000 years. ” Gil agrees: “There are a class of problems in the world that we cannot solve efficiently with a classic computer. Not now, not 20 billion years from now, not ever. We are not saying that quantum is going to solve all the problems that are difficult, but that it is the only technology that alters what is possible to solve ”.
As he speaks, Gil walks the halls of the impressive IBM Research building in Yorktown Heights, an hour and a half drive from New York. This gigantic mass, on the back of a hill surrounded by forests, has little to do with Google's small laboratory, 4,700 kilometers away. IBM also boasts of its figures compared to its competitor: 109 years of history, six Nobel laureates, 4,000 people working in R&D worldwide (although the company does not clarify how many of them do it in its quantum division). It has 16 quantum systems in the cloud with 220,000 users from more than 100 organizations that have written more than 225 scientific papers with their connected systems. And a machine that it introduced last year, Q System One, and that is going to be installed in Japan, as part of the collaboration with the University of Tokyo, and in Germany, with the Fraunhofer Institute.
Across the country, Boixo boasts Google's quantum supremacy, an achievement never before achieved. "We started doing calculations three years ago," he explains. “The idea was to demonstrate that there really is a different method of computing and that in practice it does work. It seems to us a very important scientific milestone in the history of computing. ”
"Quantum computing is going to be a reality and it is going to affect our lives. Maybe sooner than we think"
SERGIO BOIXO, scientific head of quantum computing theory at Google
None of the IBM figures, nor Google's achievement, impress Juan Ignacio Cirac too much. The Spanish researcher, director of the Max Planck Institute for Quantum Optics, is one of the most important personalities of quantum science in the world. He reflects: “It is an interesting moment, very interesting, but what we have is not yet quantum computing. That moment, which we know will have a huge impact on society, is still a long way off. " And he explains why: “If you have visited the laboratories of Google or IBM, you will have seen that their chips have many cables only to handle around 50 cubits. The key is to go from 50 cubits to 50 million. And that moment is far away. "
For that quantum race car to reach the finish line when the starting gun goes off, it has yet to overcome a difficult road. As Cirac explains, the maximum number of cubits currently reached by the IBM and Google chips is around 50. But that amount is still minuscule. And researchers cannot simply increase it. Quantum chips are extremely delicate. They are in very controlled, isolated laboratories, surrounded by the most complex technology to make them work. Antonio Córcoles, a researcher at the quantum team at IBM Research in Yorktown Heights, explains in the laboratory what is happening in that beautiful chandelier surrounded by cables: the superconducting cubes work with microwaves and must be cooled, since all heat in the system is translates into noise that can cause errors. The quantum processor is at the bottom, which is the coldest. This part is about 250 times colder than outer space, and this temperature is reached progressively over several hours from room temperature. Hence the size of the system and the number of cables that support it. All this means that you cannot increase the number of cubits of your computer like crazy; the problem is to do it without also brutally increasing its size and the number of cables connected to it, maintaining stability and without increasing errors, because a computer with errors, however quantum it may be, is useless.
In Google's laboratory, Boixo agrees: “To go from 50 cubits to a million there are many very difficult problems to solve. We want to continue increasing that number, but the level of errors has to go down ”. So quantum chips still do very simple tasks; They are the babies of the real computers that, according to the experts consulted, could take between 10 and 30 years to arrive. "In technology it is very difficult to predict anything beyond 10 years," explains Juani Bermejo, who is a researcher Athenea3i-Marie (Sklodowska) Curie in quantum computing. "But predictions of when quantum computers will be useful require error correction technology to be developed that has not yet been developed." Bermejo insists: Quantum computers are not flying or close to flying, in the Wright brothers' metaphor that Google uses. "They are in diapers."
Thirty years is a lifetime for an individual. But for big companies, research centers and governments it means that the time to prepare for the great quantum era is now. Federico Carminati is the director of innovation at CERN, the European Organization for Nuclear Research, the largest research laboratory for particle physics in the world and the closest thing to a cathedral of science, if science believed in God. CERN already works in quantum computing with IBM, Google, Intel or Microsoft, and Carminati explains why: “For me it is like buying a lottery ticket. If you ask me if quantum computing will come, my answer is yes. The question is when. But, in the meantime, it is very important that at least a few people in the organization understand and know what quantum computing is and how to use it. If it doesn't come, it will be an interesting intellectual adventure. And if it arrives, we will be prepared to use it and exploit it ”. Katie Pizzolato, director of IBM's network of quantum technology clients, explains that the company's work with those clients who already use their quantum technology in the cloud (pharmaceutical companies, automotive companies, financial companies or CERN itself) is something different from usual: “They tell us: 'These are the questions that could be interesting, valuable or disruptive in our industry. How can we begin to face these problems from a quantum perspective? ' It's been a long time since we've had a blank whiteboard in technology, so what we're asking is what would you want to do now that you can't do with current technology. And we started working. " For Ismael Faro, head of software and cloud services for IBM Quantum in Yorktown Heights, the best thing about working day after day in a method that is not yet clear what it will do is "share time with people who have a passion for technologies that only a few years ago they sounded like science fiction; it is like a window to the future ”.
PODCAST Extra EPS: Quantum Power
Companies are investing a lot of money in this technology, which is also very expensive. Neither Google nor IBM make public the figures of what they spend on it, nor the customers who use it count how much they spend, but one fact is enough to get an idea: the 2000Q quantum prototype of the company D-Wave costs 15 million dollars ( just over 13 million euros). Venture capital firms invested around $ 1 billion in newly created startups in 2017, according to Statista data. 925 patents related to quantum technology were also registered that year, double the previous year, according to the same source, which shows how the interest of different companies and companies in this field has increased. The interest has also caused the search for quantum professionals to intensify, which do not yet exist in the quantity claimed: "In some companies they have hired students without a doctorate," Bermejo explains. “It is formed relatively slowly, also because there are not so many university professors who know about this; Until recently, the quantum were the geeks of the universities. You have to train people to train others, "he adds. Cirac says that in their training centers in Munich they begin to take masters in quantum computing, and they are also implementing them in universities such as Berkeley, Stanford, MIT or Harvard. "Professionals of the future have to be prepared in fields like physics and quantum computing. There will be new jobs related to engineering and, later, when we have quantum computers, including programming, "he explains. The Spanish Paco Martín, leader of cloud development at IBM Quantum in Yorktown Heights, recommends that young people who want to train in this technology "open their minds, not think classically while assimilating the basic concepts."
According to the BCG consultancy, the quantum business could reach 60,000 million dollars in 2035 and the governments want us to stay out of this revolution. The one who leads the way, at a great distance from the others, is China. Its laboratory for quantum information sciences will invest 10 billion dollars between 2017 and 2020. The United States announced last year a national strategy by which it will invest 1.2 billion until 2023. And in 2016 the European Union launched an initiative that will invest 1,000 million in 10 years. And it is not just about being the first to get powerful computers that make great discoveries here; it's also about hiding your current secrets from your future quantum rival.
The lamb's mother of quantum technology is crypto. Today's security systems are protected by cryptographic keys that can be more or less decipherable depending on their complexity and how skilled hackers trying to violate them are. But quantum technology completely changes the rules of the game, so you have to find a new security protocol that protects data, but not data from the future, but from today. “We know that quantum technology is capable of deciphering the encryption systems in use right now; We have known for a long time, and that also means that we have had enough time to prepare ourselves, ”explains Boixo. Gil stresses the idea: "The problem is real and you have to act now." Data storage is very cheap, so a smart hacker or a rival government can store your encrypted data and, in the future, use a quantum machine to quietly decrypt the past.
China has already announced that it wants to be the quantum leader in 2035 and is working on an impenetrable communication network that it has already done the first tests on. Chinese movements have made some U.S. senators very nervous, pushing the Donald Trump administration to make its own investments. "In the same way that atomic weapons symbolized power in the Cold War, quantum capabilities will possibly define hegemony in our increasingly digital, global, and interconnected economy," Republican Senator Will Hurd wrote in Wired magazine. Therefore, companies are calling to implement a new encryption protocol that is secure as soon as possible. "There is an urgency to change our most sensitive data to the new protocol. Every day that passes using the system of the past is another day that you leave vulnerable to decryption of the future, ”explains Gil.
That new future, the possibilities of which we are only beginning to see, may be harmed by what experts call the "quantum winter", a drop in expectations and investments dedicated to this business if the technology advances more slowly than expected, more even now that a recession of unforeseen consequences will occur due to the coronavirus pandemic. "I think there may be disappointment, and less money, when people realize that what scientists constantly say is happening, and that is that quantum computers cost a lot to build," says Bermejo. “The best way to avoid this is to keep working on practical applications and a fault-tolerant computer. We try not to hide the difficulties: we are at the beginning of this career ”, says Boixo. Cirac is more forceful: “I am convinced that the quantum winter is going to happen. Those are the 10, 15 or 20 years that we talk about until we have a computer. Right now we have what Americans call a hype about quantum computing. And then, in time, there will be a resurrection. "
This hype , which no one denies, does not hide the relevance that this technological change will have in the economy, industry and in our lives, even what Boixo hopes will be "a new industrial revolution." Until then, there are a few chores along the way, such as bug fixes that transform our baby quantum processors into true adult quantum computers. The training of a new generation of physicists, engineers and programmers will also have to be improved. And you also have to reflect on what Bermejo calls "responsible quantum research, because technology, if not done in a human way, can end badly." The idea is that it does not happen as with artificial intelligence, weighed down by the absence of that human factor that explains why it has so many biases and often lacks ethics. But whatever the obstacles, and no matter how long it takes to avoid them, all experts trust that this new way of solving problems will lead us to make discoveries and achieve achievements that our limited Homo sapiens mind, which barely understands what's going on in the little quantum underworld, he can't even begin to imagine. — eps