The Era of Quantum Utility

What’s changed here is that while quantum computers had tremendous promise and theoretical opportunity was evident, they are inherently noisy. These turn out to be rather delicate quantum states, these simultaneous states of 0 and 1 that give this tremendous computational power. Any kind of noise or interference can create errors that hamper the machine’s performance. 

Recently, we’ve been able to quiet the qubits, our hardware has improved, and our ability to mitigate errors in the hardware — those errors that remained — has dramatically improved. We published at IBM, on the cover of Nature, our breakthrough evidence that current quantum computers do show very strong evidence of utility and can actually perform calculations that even a supercomputer cannot perform. That’s an extraordinary moment in quantum computing, where this theoretical promise that’s been out there for decades is finally being evidenced in a real way with something that is very useful.

Ihrie: Is that threshold the point where you can show there’s a computational “advantage” to the quantum computer versus a classical computer?

Broz: We prefer to call it “computational utility” at the moment, but it is a form of advantage. We took one of our systems and performed a calculation on it and compared that to the same calculation being run on a supercomputer by a team of computer scientists at the University of California, Berkeley, and at Lawrence Berkeley National Laboratory, who simulated the same model we were running on the quantum computer. And as the problem became more and more complex, the quantum computer continued to perform accurately, but the classical supercomputer system eventually faltered.

Ihrie: IBM is clearly a leader in development of qubit structures. There’s a fair amount of discussion about other approaches that are  being called quantum computing, which I  would classify as at least taking advantage of some of the quantum properties of say, photons, in nonclassical computing architectures. Do you have any insights into that range of developments?

Broz: We have many worthy competitors. Some really good people are in this field, and we have some excellent competition, which in the American way only makes us better. I think we offer unmatched accessibility, reliability, and scalability of our devices, and this is our very significant advantage as a company.

There are other qubit modalities. Other people in the competitive space are using the same type of qubit we are. People are using spin qubits, and there are those looking at photonic devices, and those that use trapped ions. All are a bit different from each other, and all are making progress in their own ways. But IBM is, I think, the indisputable quantum industry leader at this point. We were in the cloud in 2016. We have 25 quantum computing systems in the cloud today. We have over half a million registered users on our system. We run a couple billion circuits a day. Over six million people have been trained on our software globally. IBM Quantum is a success story.

We’ve always approached this as a full-stack solution. We go from the hardware to the software, the middleware, all the way to the application. We make sure we can scale every layer of that stack and make it addressable, accessible, democratized technology for users so that this becomes a technology that can be used by anyone — not just quantum physicists in an esoteric setting.

Ihrie: You mentioned the capability’s cloud hosting. Will this modality be more like the supercomputer centers or cloud-accessible compute capability, as opposed to an enterprise-level or a desktop-level capability?

Broz: I think we’ll see this as quantum data centers — quantum-centric supercomputers. But we are pushing toward our extremely large systems. We’ve announced plans to put in place, within the next decade, a 100,000-qubit machine. That’s a very ambitious goal, but it’s one that we feel very confident in achieving. We think we have the partners and the team that can perform the research and the very hard work needed to get there. As we scale this up, we’re scaling along with it our capabilities, our software, our ability to utilize these machines, and the applications for these machines. 

There are challenges we need to overcome. It’s just a lot of hard work and not some mysterious process that needs to be invented. Difficult research is yet to be done, and some leaps to be made, but nothing that would require an absolute miracle we can’t get our arms around.

We’re partnered with the University of Chicago and their laboratories that they run for the Department of Energy. We have great partners with Oak Ridge, Argonne National Laboratory, Fermi Laboratory, and Lawrence Berkeley National Laboratory. We’re also partnered with the University of Tokyo and RIKEN in Japan on applications. So, we really are blessed with very, very good partners globally and a great team internally.

Ihrie: We’re trying to position the Commonwealth of Virginia to take advantage of quantum computing. Virginia is a leader in things like cybersecurity because of the presence of Amazon and a number of others. Logistics is clearly an area of interest. How do we start preparing a workforce of people capable of taking advantage of these capabilities?

Broz: A vibrant ecosystem for quantum really comprises four elements. One of those, as you mentioned, is the workforce. And I know several schools in Virginia are making efforts to add quantum information science into their curriculum. A number of institutions in and around the DMV area are putting quantum information science in, or have had quantum computing in, their curriculum for quite some time. 

The second would be the research and academic communities. The third would be industry and making sure that Virginia industries are engaged in quantum, particularly Northern Virginia industries around the defense and aerospace community. The fourth element in the ecosystem would be access to actual quantum computer systems. I’d love to see Virginia follow suit with New York, Ohio, Japan, Quebec, Germany, Spain, and other areas of the world where they’ve taken the step to put a quantum system on premises. 

Until that time, we offer a great opportunity for companies, organizations, and universities to access our advanced quantum fleet through the cloud. Through either one of those modalities, I think access is super important. Developing that workforce, using that access, beginning to explore industry use cases, and then training the workforce around that is all a very important part of a synergistic ecosystem.

Ihrie: In terms of your roadmap, how do you see the future going? 

Broz: I think it’s here and now. The writer William Gibson said that the future is already here — it’s just not evenly distributed. At IBM, we’ve made it clear that this is the quantum decade. Our mission, our goal, is to bring useful quantum computing to the world. We’ve had a breakthrough in error mitigation research and we really are in an era of quantum utility. Even though quantum systems are noisy, we now have tools in our toolbox that allow it to be applied to a range of important and difficult problems that might even be impossible for classical supercomputers, but are addressable by quantum computers. I think, once realized, quantum-centric supercomputing will open up new, large, powerful computational spaces we can’t even imagine today. 

For the full interview, visit www.vedp.org/Podcasts