What’s next for quantum computing

What’s next for quantum computing


This story is part of MIT Technology Review.The Next Series:We look across industries, technologies, and trends to give you a glimpse into the future.

In 2023, quantum computing’s progress will not be measured by major hardware announcements. Instead, it will be determined by years of hard work by researchers, who get chips to talk to each other and shift away from trying to deal with noise as the field becomes more global.

The headlines about record-setting systems dominated the news cycle for quantum computing for years. Researchers at IBM and Google Have had spats It was a matter of opinion as to who did what and whether the effort was worthwhile. The time for arguing about who has the largest processor seems to be over: the firms are now focused on the real world and are preparing for it. Everyone is acting like grown-ups suddenly.

To emphasize just how eager researchers are to get off the hype train IBM is expected announce a processor in 2023 This is in direct contravention to the trend of adding more quantum bits (or “qubits”) to the mix. Qubits, which are the processing units of quantum computer, can be made from a variety technologies, including superconducting circuitry and trapped ions. Photons, which are the quantum particles of Light, can also be used to build qubits.

IBM has been interested in superconducting qubits for a long time. Over the years, the company has made steady progress in increasing the number of qubits it can pack onto a chip. IBM, for instance, unveiled a record-breaking number of 127 in 2021. It debuted its November launch. 433-qubit Osprey processorCondor, a 1,121-qubit processor, is the company’s goal. Condor will be released in 2023.

IBM will also debut its Heron processor this year, which will only have 133 qubits. Although it might seem like a backwards move, Heron’s qubits are of the highest quality. Each chip will be able connect directly to other Heron processors. This is a significant shift from single quantum computing chips to “modular” quantum computer built from multiple processors connected together. This move is expected to increase quantum computers’ scale.

Heron is a sign of bigger shifts in quantum computing industry. Experts suggest that we may see general-purpose quantum computer sooner than many expected due to recent breakthroughs, aggressive planning, and high levels funding. Michele Mosca is the deputy director of the Institute for Quantum Computing, University of Waterloo.

Here are some areas experts expect to see improvement.

Stringing together quantum computers

The IBM Heron project is just one step in the world of modular quantum computing. The chips will be connected to conventional electronics so that they can’t maintain the “quantumness of information” as it moves from one processor to another. However, the hope is that these chips, when connected with quantum-friendly microwave or fiber-optic connections, will allow for distributed, large-scale quantum computing with up to a million connected qubits. This may be the number of qubits needed to run error-corrected, useful quantum algorithms. “We need technologies that scale both cost and size, so modularity is important,” Jerry Chow, director of IBM Quantum Hardware System Development, says.

Similar experiments are being conducted by other companies. Peter Shadbolt is chief scientific officer at. PsiQuantum, which uses photons to power its qubits. PsiQuantum is finishing up a silicon-based modular chips. Shadbolt claims that the last piece it needs–an extremely fast and low-loss optical switch –will be fully demonstrated by 2023. He says, “That gives us an entirely feature-complete chip.” The warehouse-scale construction can then begin: “We’ll take all the silicon chips we’re making, and assemble them in what will be a building scale, high-performance computer system.”

According to, the desire to shuttle qubits between processors is a sign that a neglected quantum technology will soon be in demand. Jack HidarySandboxAQ CEO, a quantum technology company, was Last year, Alphabet spun off. He says quantum communications, in which coherent qubits can be transferred over distances as high as hundreds of kilometers, will be an integral part of the quantum computing story of 2023.

Hidary stated that the only way to scale quantum computing is to create modules with a few thousand qubits, and then link them to achieve coherent linkage. It could be in the same place, but it could also be spread across campus or across cities. Distributed computing is well-known in the classical world. However, for quantum, we need to have coherent links. This could be a fiber-optic network with repeaters or some fiber that connects to a ground station and a network of satellites.

In recent years, many of these communication components were demonstrated. For example, China’s Micius satellite demonstrated that coherent quantum communications can be achieved between nodes separated by 1,200 km. In March 2022, an international team of industrial and academic researchers was established. Demonstrated A quantum repeater that relayed quantum information over 600 kilometers fiber optics.

Take on the noise

The industry is establishing qubit networks, but it is also abandoning an idea that was popularized in the last five year: that chips with only a few hundred qubits could be capable of useful computing, even though noise can easily disrupt their operations.

This idea, known as “noisy intermediate quantum” (NISQ), was a way to get some short-term benefits out of quantum computing. It could have been years before we reach the ideal large-scale quantum computer with hundreds of thousands of qubits dedicated to correcting errors. However, NISQ optimism seems to be waning. “The hope was that these computers would be used well before error correction could be done, but the emphasis is shifting away,” says Joe Fitzsimons CEO of Horizon Quantum Computing, a Singapore-based company.

Some companies are attempting to improve the traditional form of error correction by using some qubits to correct others’ errors. Both Google Quantum AI, and Quantinuum, a new company founded by Honeywell Quantum Computing and Cambridge Quantum Computing. issued Papers This demonstration shows that qubits can be combined into error-correcting ensembles which outperform the underlying qubits.

Other teams are also trying to figure out how to make quantum computers fault-tolerant without as much overhead. IBM, for instance, has been studying the error-inducing noises in its machines and then programming to subtract it (similarly to noise-canceling headphones). It’s not perfect, however. The algorithm is based on a prediction of what noise will occur, not what actually occurs. Chow states that it does a decent job. “We can build an error correcting code with a lower resource cost, which makes error correction feasible in the near future.”

Maryland-based IonQSimilar work is being done by, which is creating trapped-ion quantum computer. Chris Monroe, chief scientist at IonQ, says that most of our errors are caused by us. “We poke at the ions and run programmes.” “That noise is known, and we have used different mitigation methods to push our numbers.”

Software: Get serious

Many researchers believe that programming needs to be given more attention despite all the hardware advancements. Michal Stechly, of Michal Stechly says that “our toolbox is definitely limited” compared to the tools we will need in 10 years. Zapata Computing, a Boston-based quantum software company.

Circuit-based computing is the way code runs on cloud-accessible quantum computers. This means that the data is run through a predetermined series of quantum operations before the final quantum measurement is made. The output is then given to the user. Fitzsimons states that this is a problem for algorithm designers. Conventional programming routines involve looping steps until the desired output is reached and then moving on to another subroutine. Circuit-based quantum computing is a method of computing that produces an output. There is no way to go around.

Horizon Quantum Computing is one company that has been developing programming tools to allow flexible computation routines. Fitzsimons states that this gives you access to a new regime in terms the types of things you can run. We’ll be rolling out early access in 2012, Fitzsimons adds.

Helsinki-based Algorithmiq It is also innovating in programming. CEO Sabrina Maniscalco says, “We need nonstandard programming frameworks to program current quantum devices.” Algorithmiq’s new drug discovery platform Aurora, which combines quantum computation results with classical algorithms, is now available. This hybrid quantum computing is a rapidly growing area and is widely recognized as the way the field will function in the long-term. According to the company, it hopes to demonstrate that a quantum system can perform better than a classical computer in real-world, relevant calculations by 2023.

Global competition

There is also likely to be change on the policy front. Representatives from the government, including Alan Estevez, US Undersecretary of Commerce for Industry and Security, have participated in this meeting. Hinted Trade restrictions regarding quantum technologies are on the horizon.

Tony Uttley is Quantinuum’s COO. He says he is actively in dialogue with the US government to ensure that this doesn’t negatively affect an industry that is still young. He says that about 80% of the system’s components and subsystems are made from components or subsystems purchased from outside the US. “Putting a limit on them doesn’t help, and it’s not something we want to do when we compete with other companies in other parts of the world.”

There are many competitors. Baidu, a Chinese search company, opened access to a number of competitors last year. 10-superconducting-qubit processor It hopes that this will allow researchers to explore quantum computing in areas such as materials design and pharmaceutical research. According to the company, it has just completed the design of a 36-qubit ultraconducting quantum chip. A spokesperson for Baidu told MIT Technology Review that the company will continue to make breakthroughs in integrating both quantum software and hardware and help with the industrialization and commercialization of quantum computing. Alibaba is also home to researchers who are working on quantum computing using superconducting qubits.

Fujitsu Japan is collaborating with Riken research institute to provide companies access to Japan’s first domestic quantum computer. This will be available in the fiscal year that begins April 2023. It will be equipped with 64 superconducting qubits. Shintaro Sato (head of the quantum laboratory at Fujitsu Research) says that the initial focus will be on materials development, drug discovery and finance.

However, not everyone follows the well-trodden path of superconducting. The Indian government will be announcing its 2020 plans in 2020 Promised to spend 80 billion rupees Quantum technologies will receive $11.22 billion (at the time of the announcement). Photonics technologies will receive a large chunk, including for satellite-based quantum communications and innovative “qudit” photonics computing.

Qudits increase the data encoding capabilities of qubits. They offer three, four or more dimensions as opposed to the binary 0 and 1. However, this does not necessarily increase the possibility for errors to occur. Urbasi Sinha, who is responsible for the quantum information and computing laboratory at Bangalore’s Raman Research Institute, India, says, “This is the kinda work that will allow me to create a niche rather than competing with the existing work for many decades.”

Although things are becoming more serious and internationally competitive in quantum technology, the field is still largely collaborative for now. Monroe states that while competition is fierce in this field, it is also something Monroe likes. “We don’t have a zero sum game mentality. There are many technologies out there at different maturity levels, and we all play together right here. There will be some consolidation at some point, but not yet.

Michael Brooks is a freelance science journalist, based in the UK.

Continue reading