An Interview with the Founder and CEO of Rigetti Computing
While quantum computing has attracted a number of startups, most tend to focus on a particular aspect of the technology challenge or operate in one layer of the quantum tech stack, leaving R&D on end-to-end quantum computing to the tech giants. Chad Rigetti’s eponymous Berkeley, California–based company is dedicated to developing the full quantum computing stack based on superconducting qubits technology, and as such it is competing head to head with the likes of Google, IBM, and Intel.
Equipped with a PhD in applied physics from Yale, Chad formed the company in 2013 after spending three years at IBM’s quantum computing labs. He has since raised almost $120 million in venture capital. In connection with the BCG report, The Next Decade in Quantum Computing and How to Play, BCG senior partner Philipp Gerbert, talked with Chad about his vision for quantum computing and how he sees the technology developing.
How did you get into quantum computing?
When I was in college, two things converged. I got frustrated by a course on quantum mechanics—I was having a lot of trouble understanding what it was trying to teach me. At the same time, I was fascinated by a course on computer science, particularly how computers store, process, and present information. As I began to grasp quantum physics, I put the two things together. Quantum computing was a way to pursue both interests. Then I found Michel Devoret and his group at Yale building superconducting circuits on silicon chips. I successfully applied as a PhD candidate, and Michel accepted me into his program.
I also grew up on a farm. I’ve always built things. Maybe it was inevitable that I’d want to build a quantum computer. It’s the thing that’s animated me for the last 20 years.
Quantum computing has been hailed as the inevitable future. Do you agree?
Yes, quantum computing is to the 21st century what microelectronics or semiconductor technology were to the second half of the 20th century. It will drive national competitiveness and economic markets. The problem for classical computing is that Moore’s law has reached the point of being an expensive model: classical chips have become a high-capital commodity with diminishing returns. Quantum computing shifts the traditional scaling of computer power. It also changes the game from commoditized hardware to defensible, advantaged hardware.
From a market perspective, I think many different modalities of quantum computing are inevitable and will create value. Different technologies will develop on different time scales, and each will find the applications and uses to which it is best suited. Superconducting is an integral part of quantum chip technology, for example. Other technologies will be more easily used for CPUs, RAM, and storage, all of which still need to be developed over the next 20 to 30 years.
You mention superconducting. Why are you pursuing that over other technologies?
With different technologies, everyone has a religion, and it becomes a contest about which one is in the lead, or which one is best. I think that’s the wrong way to look at it. Superconducting is best suited for building scalable quantum processing hardware, in my judgment, and it’s the technology that’s shown the most progress most quickly in recent years. I think most people agree that superconducting will be the first to produce qubits that achieve quantum advantage. That doesn’t mean that ion traps or some other technology won’t have a big breakthrough down the road. But right now, superconducting has a huge leg up: it’s the only quantum computing technology that is based on chips and lithography-driven. That means that once we solve the error correction challenges, it’s relatively easy to scale, which is the key to achieving quantum advantage.
Many people talk about the long term in quantum computing. But what about the next few years, the first applications and use cases?
I look at the quantum computing market as having three phases. The first, which we recently moved out of, was the what-is-possible phase. People could see the potential, but the big question was, can we build a programmable quantum computer? We’ve answered that question, yes, we’ve shown it can be done. We’re now in the second phase, the early market phase. We know the machines are real, but nobody has actually built one and used it to solve a problem that is not also solvable using classical computers. When we do that—which is my definition of quantum advantage—we’ll move into the third phase, the growth phase for quantum computing. That will be defined by the development of new markets and domains that are rooted in quantum advantage.
What do you think is exaggerated about quantum computing in the current public debate? What is underestimated?
I think the potential is such that most people don’t exaggerate, but some do lack appreciation for how much work needs to be done, at all levels, on all aspects. For example, to move to phase three, the growth phase, we not only need more powerful machines with more qubits, but the development of integrated quantum computing systems with programming languages, software development kits, algorithms, and so forth. There’s an entire ecosystem that has to come into formation.
You’ve offered a $1 million prize for proving quantum advantage. What was your thinking?
Quantum advantage will be quantum computing’s Wright Brothers–Kitty Hawk moment. It will demonstrate market viability, it will jump start the ecosystem development I was just talking about. Rigetti Computing is one company. We can’t do all of it. Everyone in the ecosystem will benefit from the success of others.
What advice would you give to business executives? How should traditional companies get involved?
C-suite executives need to ask themselves three questions about their quantum computing ambitions. What do I want to achieve? Do I want to get involved and start building a capability? Do I want to be an owner of the disruption that the technology will inevitably generate? The answers will dictate what your quantum computing strategy should be.