Amazon joins quantum race with ‘cat-qubit’ powered chip

Amazon has become the third tech giant in as many months to announce a breakthrough in quantum computing – a technology that promises vast processing power but is beset by technical difficulties.

The firm has unveiled Ocelot a prototype chip built on “cat qubit” technology – an approach that derives its name from the famous “Schrödinger’s cat” thought experiment.

The chip seeks to address one of the biggest stumbling blocks to the development of quantum computers – making them error free.

Amazon says, taken alongside other recent breakthroughs in the industry, its work means useful quantum computers are likely to be with us sooner than previously thought.

But how quickly these machines will be powerful enough to be practically useful for a range of commercial applications is a matter of debate among experts.

Oskar Painter of Amazon Web Services (AWS) Center for Quantum Computing at the California Institute of Technology, where the work was carried out, told the BBC that recent progress meant an “aggressive date” of a decade was now “looking more and more realistic”.

“Five years ago I would have said maybe 20 or 30 years”, he said but added “this timeline’s come in quite a bit.”

Ultimately AWS, which provides cloud computing services, would like to offer quantum computing services to its customers, but Mr Painter also said he believed the advanced machines could eventually help optimise the vast global logistics of Amazon’s retail business.

“You know, a company like Amazon, you make a one percent improvement in that and you’re talking large dollars right? Quantum computers could enable you to do that more effectively, more real time – and that’s the real value there,” he explained.

Quantum computers solve problems by exploiting the strange properties of matter and energy at very small scales, as described by the science of quantum physics.

Quantum computers won’t replace so-called “classical” computers, but promise to be able to solve problems even the most powerful modern computers cannot – yielding new discoveries such as better batteries and new medicines.

But that potential is being held up by the problem of errors.

Quantum computers are extremely sensitive to noise in their environment – vibrations, heat, electromagnetic interference from mobile phones and WIFI networks, or even cosmic rays and radiation from outer space can all cause them to make errors, which then need to be corrected.

Cat qubits are one attempt at solving this problem by engineering error resistance into the design of the qubits it uses.

Qubits are the fundamental elements of quantum computers, the equivalent of bits in the computers most of us use today.

Cat qubits are named in honour of Erwin Schrödinger, whose cat-in-a-box question in 1935 helped illuminate some of the thinking behind quantum theory.

Amazon believes the new chip, which has just five cat qubits out of a total of 14 key components, could reduce the costs of correcting quantum errors by up to 90%, compared to current approaches.

The technology of cat qubits isn’t exclusive to Amazon, a French company named Alice & Bob carried out pioneering work on the tech and continues to evolve the technology.

Mazyar Mirrahimi director of research at the French national technology research institute, Inria said Ocelot was “an important step forward towards hardware-efficient fault-tolerant quantum computation”.

Amazon believes the new chip offers a path to scaling up to more powerful machines with this type of error proofing built in, but researchers admit there are many challenges ahead.

Michael Cuthbert director of the UK’s National Quantum Computing Centre welcomed the progress Amazon had made but told the BBC it remained to be seen the effect it would have on the speed with which the industry is able to develop really useful quantum computers:

“Error correction is a vital step necessary in the long-term development of quantum computing. It is the crucial step that turns quantum computing into a practical and commercial tool we can use to solve complex problems in chemistry, materials science, medicine, logistics and energy.”

“Part of the challenge is how to scale the revolutionary technology efficiently – mechanisms that enable error correction without huge overheads in chip size, energy consumption and systems complexity are really welcome.”

Amazon researchers have published their findings in a research paper in the scientific journal Nature

Amazon joins Microsoft and Google in announcing a new experimental chip. But is this flurry of announcements the result of clever research or clever PR? Or is it coincidence, the tech equivalent of the saying buses always come in threes?

Heather West follows the quantum computing industry as a research manager for the International Data Corporation and was briefed on the new chip by Amazon in advance of publication.

She describes Amazon’s results as an “advancement” rather than a breakthrough.

All three recent announcements have focused on reducing errors, and she tells me the industry is “pivoting” from a focus on the number of qubits to a focus the “ability to use these these systems at scale to solve real world life problems. And by doing so we need to be able to solve the error correction within the quantum systems”.

However Mr Painter agreed “100%” that it wouldn’t be easy to scale up today’s experimental systems.