When we first started the Quantum Computing Report in 2015 there were disagreements within the quantum community on whether a quantum computer would ever be able to do anything useful. Since then much progress has been made and there are only a handful of doubters left that believe we will never see a commercially viable quantum computer.
Now almost everyone agrees that we will have quantum computers doing productive work in the future. But the question remains when? The disagreements we hear today in the quantum community is whether the NISQ era computers will be able to solve commercially valuable problems or whether we will need to wait for a fully error corrected machine with millions of qubits before we can do anything truly useful. Even Professor John Preskill of Caltech who coined the term NISQ (Noisy Intermediate Scale Quantum) had this to say at the Q2B Conference last December. “NISQ will be an interesting tool for exploring physics. It might also have other useful applications. But we’re not sure about that.”
This is an important question for company developing quantum hardware and quantum applications. We see some companies including IBM, Rigetti, and many others promoting the concept that there will be some early real world applications that will be possible with machines that have a few hundred or so non-error-corrected qubits. We’ve also seen specific hybrid classical-quantum algorithms such as VQE (Variational Quantum Eigensolver) and QAOA (Quantum Approximate Optimization Algorithm) developed for optimal use on a NISQ machine because they use short quantum circuit depths and the algorithms can recover even if there is an error in portions of the computation.
On the other hand, we have seen other companies such as Intel and PsiQuantum signaling that they are focusing their development efforts on a fully error corrected machine. By bypassing the creation of a NISQ version and focusing their efforts on the more advanced architecture, they may be able to accelerate their introduction of an error corrected quantum computer and leapfrog over the competition.
It’s our view that NISQ computers will indeed be able to show commercial usefulness and developing them is a worthwhile activity. We see a lot of innovative engineers and programmers in the quantum community and they have the potential to develop creative solutions for problems even when there are deficiencies in the hardware. We also believe that there is great value in getting an early version of a product in the hands of an end user to get feedback. The VC’s have a term for this called MVP – Minimum Viable Product. In many cases, an end user will point out problems or suggest improvements that would not have occurred to the company that is developing the product. Finally, we see introduction of early commercial applications using a NISQ quantum machine would be a great way to prevent a Quantum Winter. Although a lot of VC and government investments are being made right now, these investors will want to see measurable results within the next few years. If that doesn’t happen, they may become disillusioned and the investment money will dry up.
So, like many things, there is a great diversity in approaches, technologies, and strategies for quantum computing. But overall, this diversity and competition is a strength for the industry overall and will enable success as the industry develops.
February 13, 2020