According to a recent report published in the journal Nature, researchers and scientists at Google claim that a quantum computer performed a computational task that would be impossible for a conventional computer to complete. Thus, it demonstrated what is known as “quantum supremacy” for the first time ever.
According to the report, Google’s 53-qubit Sycamore quantum computing machine was able to finish a computing task in 200 seconds. This was significantly faster than the 10,000 years it would take the world’s most powerful supercomputers to accomplish the same task. However, as Google acknowledges, the task involved executing a randomly chosen sequence of instructions, rather than solving a real-world problem. And that’s why despite the optimism, many in the quantum computing community view the achievement as a milestone along a very long journey to building quantum computers that can effectively and efficiently address specific real-world use cases.
So, does Google’s announcement signal a new era for quantum computing? Does it herald an era marked by more frequent technological achievements that will generate headlines in and beyond the technology press? It may be too soon to tell, but John Preskill, the physicist at Caltech who coined the term “quantum supremacy” itself 7 years ago, is somewhat circumspect. Preskill told the Los Angeles Times that the announcement “won’t change anything overnight, but it is significant that quantum computers are now at the stage that at least in some arena, they can outperform the best computers on Earth.”
Of course, in the highly competitive quantum computing marketplace, Google’s competitors have expressed skepticism over the announcement. The Google researchers utilized IBM’s supercomputer to benchmark its quantum computer, which led to the claim that a classical machine would require 10,000 years to perform the random number error validation computation. However, according to a blog post, IBM said that Google never optimized the IBM supercomputer prior to running the test. As per the blog, if properly tuned, the IBM supercomputer would’ve been able to solve the problem in just 2.5 days, thereby muting some of the enthusiasm surrounding Google’s announcement.
Regardless of IBM throwing water on Google’s work, a few key takeaways from the announcement are worth noting. First, there is still some debate as to what was actually proven or demonstrated by Google. Achieving quantum supremacy in a laboratory environment, focused on a very specific problem, is just a baby step on the way to being able to demonstrate quantum supremacy on other types of problems or calculations. While the researchers and scientists at Google should be commended for their efforts, a massive amount of work remains to be done in terms of developing quantum machines that can solve increasingly more complex problems. Further, it may not be possible to leverage or scale the knowledge gained in this initial solution to other problems. Solving random numerical problems is quite different than, say, designing a quantum machine that can calculate the millions of permutations in an optimization problem.
What Does It Mean for Commercial Deployments?
That leads to another key point. Google’s Sycamore computer is based on superconducting quantum technology. This is one approach to harnessing the power of qubits, or the tiny particles that can exist in multiple states at once, making them theoretically more powerful than the traditional electronic bits used in classical computers. But there are several different technical approaches to quantum computing being developed, including the use of trapped ions, topological qubits, and annealing technology, from a wide range of well-funded startups and legacy technology companies. It is simply unrealistic to expect that scientists from IBM, Intel, D-Wave Systems, Microsoft, Rigetti, and many others will abandon their specific approaches to developing a viable quantum computer based on a single announcement from Google.
Further, like any nascent technology, years of refinement and significant financial investment into research are necessary before quantum computing will be viable in commercial enterprises. After speaking with quantum computing vendors and commercial enterprises, Tractica believes that regardless of any progress made around the concept of quantum supremacy, most of the commercial activity in the next 5 to 10 years will be related to developing quantum-ready algorithms, as well as testing potential use cases. Quantum simulators and quantum-like processors will be used extensively. Any true quantum computing technology that is deployed in the next few years will be used as an accelerator to support traditional computing systems during very specific and narrow tasks.
Additional insight into the quantum computing market, as well as a market forecast covering the 2018-2030 time period, can be found in Tractica’s report, Quantum Computing for Enterprise Markets.