Get best of the week

Alphabet has a second, secret group of quantum computer developers

Google's parent company touted quantum superiority last year. However, she says nothing about X's secret lab, where another group of developers writes programs for quantum computers



Alphabet director Sundar Pichai touted the company's quantum computing achievements on blogs and social networks

In October, Google noted a breakthrough that director Sundar Pichai compared to Wright’s first flight. Researchers at a company in Santa Barbara, California, located 500 km from Googleplex , have achieved quantum superiority - the moment when a quantum computer performs calculations that are inaccessible to any of the usual.

This was both a remarkable event in science and an opportunity for Google to break ahead in the competition of large technology companies, where participants such as IBM and Microsoft are trying to give us the crazy new opportunities promised by quantum computers. Usually humble Pichai was carried away by the advertising of the moment, wrote a poston a blog, participated in such a rare event as an interview , and posted on Instagram his photo next to a sparkling machine that achieved the result.

And a month later with a little Pichai became director of the parent company Google, Alphabet. However, neither the company, nor its boss inspired by quantum success, tell practically anything about another Alphabet team working on quantum computing in their secret laboratory X.

In X, formerly known as Google X, they are developing breakthrough technologies that can spawn new Google business. A local small group of quantum computing researchers is not creating their own quantum computers. Its leader is more interested in creating new algorithms and applications for running on quantum computers, and creating software libraries that allow ordinary programmers to use exotic machines.

“Iron is very interesting, but most of the value is created by programs,” Jack Hydari, a serial entrepreneur who runs quantum research in X, said during a lecture at Carnegie Mellon University in November. He gave an example of companies like Microsoft that cost more than the manufacturers of iron, on which their products work, although it was the development of iron that originally spawned the computer industry.

Google and its competitors, in particular IBM, are investing in quantum computing because they believe that they can become a catalyst in major breakthroughs in many fields of science and industry, from drug development to artificial intelligence.

Quantum computers are made on the basis of special devices, qubits, which encode data into quantum-mechanical processes that occur only under carefully controlled conditions. The superconducting qubits used in the IBM and Google experiments operate at temperatures lower than in open space. Groups of qubits are capable of mathematical tricks inaccessible to ordinary computers, due to quantum phenomena that have no equivalents in everyday life - for example, quantum-mechanical objects can be "confused", after which what happens to one of them immediately affects other.

X refused to provide an interview with a specialist, both Hidari, and other employees of his team, also involved in AI research. Aisling O'Gara lab spokeswoman said the Hidari group works separately from X.

However, the Hidari autobiography published last year, where he writes about the applications of quantum computers, says he and the team work in X on quantum algorithms and software libraries. ; they work directly in building X and answer to the head of the laboratory, Astro Teller.

"He works in X, there is a small team working there," said Hartmut Niven, Google’s quantum computing project leader, when asked about the role of Hidari during an October press conference in Santa Barbara’s lab dedicated to achieving quantum excellence. “We work closely to ensure that our projects complement each other accurately.”

A neuroscientist by training, Hidari founded and listed the EarthWeb IT portal during the dot-com boomin the late 90s. In 2013, he was an independent candidate for mayor of New York on a technocentric platform, which included, as the New York magazine writes, a promise in the event of a victory "to give every inhabitant glasses Google Glasses." He received 0.3% of the vote on election day, and worked as an adviser to X at least since 2016, and in 2018 became a permanent employee of Alphabet.

X's work is based on a portfolio of ambitious projects aimed at developing breakthrough technologies, which are called “munshots,” that is, "Flights to the moon." Notable examples include robomobiles and stratospheric balloons for Internet distribution, which now stand out in separate independent Alphabet units, as well as Google Glass, the infamous wearable computer mentioned by Hidari in the election brochure. Typically, a laboratory reports on the development of its projects only after many years of work, demonstrating things that are interesting only to “nerds” like launching an Internet ball or robots sorting garbage.

O'Gara of X said that the laboratory “has no munshot projects related to quantum technology.” However, one of the members of the group, Hidari, wrote on LinkedIn that he was working on a strategy of "quantum munshots." Another expert in quantum computing from X is Juifre Vidal, recently a former full-time professor at the prestigious Perimeter Institute in Waterloo, Canada. He describes himself on Google Scholar as the principal researcher from X, and, together with Hidari and other staff members of his team, published scientific papers mentioning his link to the Alphabet lab. Perhaps this group is growing: on LinkedIn, two recruiters write about the ongoing recruitment of quantum technology experts.

And there are few such people. In a report at the Carnegie Mellon Institute, Khidari said that only 800 people around the world have a true understanding of the applicability of quantum algorithms. “We literally made a spreadsheet with a list of experts in this field. And they scored around 800 names all over the world, ”he said, and then sadly joked about the fragility of the quantum knowledge of mankind. “At conferences, we really care about the security of this community,” he said in a video report from November.

His team hosted some of these experts in November at a two-day conference on how quantum computers and tools can help physicists reconcile gravity with quantum mechanics. Speakers from X, MIT, Harvard and some Alphabet competitors in the field of quantum computers, for example, from Microsoft, were present.

Google co-founder Sergey Brin was also among the audience, according to Chris Monroe, co-founder of IonQ quantum computer startup and professor at the University of Maryland. Monroe talked about his research at the conference, among which the use of quantum computers to simulate a water molecule, but said that it was not clear what team X was specifically working on. “Nobody knows why they spend energy on quantum gravity and black holes, but this the topic intersects with quantum computers, so everything is in order, ”he says.

In a Google experiment on quantum superiority, the Sycamore quantum chip performed calculations in a few minutes, which, according to company researchers, would have taken the supercomputer 10,000 years. And although these calculations have no practical application, and the IBM quantum team disputes the statements of competitors, this result is considered a sign of growing quantum computers. Google, IBM and other companies working on iron for quantum computers are collaborating with companies such as Daimler and JPMorgan to explore how to use them. Microsoft, IBM and Amazon have announced plans or have already launched programs that allow users to test prototypes of quantum equipment through cloud services.

Despite this interest, current quantum computers are too small to do useful work, and programming for quantum computers is in its infancy. In a November report, Hidari said that the issue of facilitating work with quantum computers for ordinary programmers is critical to harnessing the potential of technology.

“Today it is such a young area that even in order to understand at least a little something about quantum computers, a huge amount of knowledge is required,” he said. “We need to provide the general public with the necessary tools.” One of the possibilities, he said, is to use machine learning in the form of auxiliary programs that adapt ordinary code to quantum systems. “Otherwise, I don’t see how to scale such an area,” he said.

A quantum team from X is trying to create more quantum experts. Hidari described a three-day quantum computer training program held at Carnegie Mellon University, which, he said, involved 600 Alphabet employees. His group also runs a graduate study program with experience in writing quantum algorithms, pays for their relocation and housing so they can work in the spacious laboratories of the former shopping center, located near the main Google campus. The hiring announcement says that new employees of the “quantum team X” will be able to run algorithms on “real quantum computers” supposedly owned by Google.

More articles:


All Articles