Although we’re still on the approach to the on-ramp of a quantum computing infrastructure, many expect 2021 will be consequential in a number of areas. These areas include machines with more power, quantum networks that extend over greater distances, and progress toward post-quantum cryptography (PQC, discussed later). For many, optimism is growing, but others remind us of the problems still to be solved.
The online tech publication TechRepublic offered both good and bad news about the upcoming year in a November 24, 2020, posting titled “6 experts share quantum computing predictions for 2021.” Those asked for comments included quantum tech executives, researchers, and industry analysts, and their educated guesses ranged widely from expect a major multiplication of use cases to watch for a pause that might result in stalled investment and interest.
ALGORITHMS AND USE CASES
The answer to the most fundamental question about quantum computing—Why quantum?—is bound to involve the incredible speed and the specific case uses for the machines. And when the president of Honeywell Quantum Solutions, Tony Uttley, was asked what he predicted in 2021 for his division at Honeywell he said, “Next year is going to be when we start seeing what algorithms are going to show the most promise in this near term era.” (See Honeywell Trapped-Ion Quantum Computer, top.)
Uttley told TechRepublic’s Jonathan Greig that quantum computers have sufficient qubits at this point to be important tools for those who know how to use them. “Next year,” he offered, “what we will see is the advancement into some areas that really start to show promise.” This opening of paths to value creation is apparently a planned objective at Honeywell.
IBM has an ambitious timeline for growth in its quantum line of computers beginning with a 127-qubit computer in 2021, a 433-qubit machine in 2022, and a 1,121-qubit computer in 2023. If reached, those machines could produce yet-unseen capabilities. (See November SF Technotes “Quantum Computing—Where Are We Now?” for details on the current state of quantum hardware and software.)
When he was asked for his predictions, Robert Sutor, vice president of IBM Quantum Strategy and Ecosystem, said “More and more people will use actual quantum hardware.” Based on reporting, “More than 130 Fortune 500 companies, academic institutions, startups, and national research labs are working with IBM to advance quantum computing through the IBM Q Network. More than 260,000 registered users are involved in the IBM Quantum Experience and it routinely executes more than 1 billion hardware circuits per day on real hardware and simulators using Qiskit [IBM’s open-source quantum computing software development kit].”
Concerning the growing shortage of quantum technicians and researchers, Sutor told Greig that IBM expects in 2021 more than 20,000 “will complete online quantum computing technology and coding courses.”
Research vice president at Gartner, Chirag Dekate, foresees a significant growth in the number of quantum service providers in 2021. He describes it as the maturation of the vendor landscape, which already includes IBM (Armonk, N.Y.), D-Wave Systems (British Columbia, Canada), Google (Mountain View, Calif.), Honeywell (Charlotte, N.C.), IonQ (College Park, Md.), Xanadu (Toronto, Canada), and many more worldwide.
For adoptive enterprises, a benefit of the increased investment in quantum and refactoring application codebases around quantum might well include the discovery of new ways of using their existing classical systems. It’s becoming clearer that classical computing isn’t going anywhere, and quantum will be taking its place alongside existing digital systems.
Overall, Dekate told TechRepublic,“Organizations engaging in these initiatives [those discovering new ways of improving their classical infrastructure with quantum] will roughly double.”
The most dramatic, and unnerving, prediction for quantum computing in 2021 came from Roger Grimes, a cybersecurity expert from KnowBe4, a leading integrated platform for security awareness. Grimes told Greig, “I predict that someone will publicly announce that they have used a quantum computer to break a traditional asymmetric key cipher. It’s been the Holy Grail since 1994 and I predict it happens next year.”
Some say the greatest threat quantum computers represent is the number-crunching ability to crack encrypted information, not over decades of attempts, but in days or even minutes. There’s a name for the event Grimes predicts and the era it will usher in. PQC will render ineffective the public-key cryptography that currently protects much of our own, our bank’s, and our government’s private information. Blockchain networks also survive on the protection of encryption.
So far, the mathematical difficulty involved in trying to crack keys and passwords has served us well. A simple randomized password like ecz5Wcx4, just eight letters and numbers, would take 9.87 years at 100,000 password attempts per second to crack, according to the Wolfram Reference app, which actually classified the password as weak. That’s using a current classical computer. The difficulty could be decreased if you used massively paralleled GPU clusters, but the end of public-key cryptography might only come with quantum computers with far more qubits than those in the current models.
Whether the threat is realized soon, or later, Grimes cautions, “Every organization and industry should be preparing for the post-quantum cryptography world.”
A QUANTUM WINTER
Another less optimistic response for 2021 came from Brian Hopkins of Forrester. He began with some skepticism explaining, “Today, nobody has convinced me any quantum computer in the world has demonstrated quantum advantage, none of them.” Then he proposed a possible “trough of disillusionment” occurring in 2021.
Like the Turing Test for classical computers, quantum advantage is a tipping point, one at which a quantum computer could demonstrate that it can compute hundreds or thousands of times faster than a classical computer. A similar but more dramatic point, quantum supremacy, is achieved once a quantum computer is powerful enough to do calculations that classical supercomputers can’t perform at all.
In October 2019, Google claimed it achieved quantum supremacy when its Sycamore processor performed a calculation in 200 seconds that they said would have taken the world’s most powerful supercomputer 10,000 years to perform. Google published its claim in the journal Nature, and IBM quickly responded on its blog that a classical system could do the same calculation (a randomness problem) in 2 ½ days, maybe fewer.
For the upcoming year, Hopkins told Greig, “[W]e’ll get a better sense of when we would achieve quantum advantage in various potential use cases. I don’t think we will reach quantum advantage in 2021, and that means in 2021, we’ll go through a bit of a trough of disillusionment. I’ve been tracking quantum computers now for a decade, and there are lots of big announcements, progress, and then it slows down.”
Hopkins’ trough of disillusionment might sound like the beginning of a “quantum winter,” but there has been enough progress in scaling the hardware and advances in quantum networks for that to seem likely. There were two AI winters during which funding and interest fell off, but those were rather extended—from 1974 to 1980 and again from 1987 to 1993. At this point, many pathways seem open, and there is substantial investment following those openings.
NOTE: On December 3, 2020, as this story was being finalized for posting, a news story appeared in the South China Morning Post, Scientific American, and ScienceNews that the Jiuzhang quantum computer in Shanghai, China, had demonstrated quantum supremacy by performing a technique called boson sampling. The computer uses light particles (photons) instead of superconductor particles like the ions in the Honeywell trapped-ion quantum computers.
According to the Scientific American coverage: “Last year, Google captured headlines when its quantum computer Sycamore took roughly three minutes to do what would take a supercomputer three days (or 10,000 years, depending on your estimation method). In their paper, the USTC [University of Science and Technology of China] team estimates that it would take the Sunway TaihuLight, the third most powerful supercomputer in the world, a staggering 2.5 billion years to perform the same calculation as Jiŭzhāng.”
Upon verification, this event would scratch the Forrester prediction from Brian Hopkins that 2021 won’t see quantum advantage (or supremacy) demonstrated.
It seems that with forecasting in this area of computing a good watchword might be, Don’t blink.