Interview transcript:
Terry Gerton Booz Allen has just announced an extension of your partnership with SEEQC to solve some big hurdles in quantum. Tell us about what you’re working on and how this partnership advances them.
Bill Vass Yeah, so this is in the quantum computing space. And so one of the biggest challenges with quantum computing is error correction. So they’re very noisy. Your current computers, like your iPhone, or your Android phone, or your laptop, whatever you happen to be using, it does have error correction code because there are literally alpha particles coming from space and flipping bits on the memory, but you don’t know it because the error correction code is there to do that, otherwise it’d be crashing all the time. But there’s a very small amount of that happening, and a very small amount of compute applied to error correction. In a quantum computer, because you’re dealing with operations on small atomic particles that you’re sort of building in the circuits of the quantum computers, basically molecules, if you want to think of it that way, are the circuits you’re putting together and you’re entangling them and operating on them in superposition. They’re very susceptible to the slightest amount of interference from the real world. So you often see them with big cryogenic stacks because they operate at very low, close to absolute zero, as close as we can get at temperatures. That’s to remove vibration. In almost all cases, they’re operating with as little vibration as possible. And any vibration is an error. Electromagnetic fields causes an error, the quality of the fabrication causes an error. So there’s a very large amount of compute applied to error correction in quantum computing. And what SEEQC is doing is working to reduce that and make it much easier to do error correction so that you can have what’s called logical qubits, or qubits that are error corrected that we can use to do amazing things as quantum computers advance. And so that, plus they’re moving a lot of the compute for error correction into the cryogenic stack rather than having to have wires running to the outside that you currently have today, which makes the quantum computer much smaller. The first machines that will come out today, over the next maybe five years or so, will be as big as a football field for one machine. So that’s not so unusual. I mean, the old ENIAC, remember the tubes, everything was big and it just gets smaller and smaller. This is the first iteration of these. And what SEEQC is doing is building not the qubits necessarily, which are the processing components of a quantum computer, they’re building everything around the qubit to make the computer smaller and more reliable and more error corrected. And so we’re writing the firmware for that at Booz Allen and that’s part of our partnership.
Terry Gerton You’ve said that this partnership will do amazing things and deliver mission-ready quantum solutions. What kind of real-world missions are you going to tackle first?
Bill Vass So the first quantum computers will have tremendous impact on what we call material sciences because a classical computer attempts to run a simulation to help design materials, but classical computers don’t work like a molecule. A quantum computer works like a molecular. So it’s really good at doing chemistry and chemical simulation. So for example, a lot of the things that we’ve invented throughout human history, we’ve accidentally discovered in the labs. I mean, from gunpowder to, you name it, we discover it, and then we’re like, oh, okay, but it would be much nicer to say, what we’d like, for example, is a high temperature superconductor, what formula would provide that? And with a classical computer, you can’t run those kinds of things, but with a quantum computer, you can. So the big impacts at first will be on chemistry, and physics, and on material sciences. And you’ll be able to start seeing effects around a hundred error-corrected qubits, which is going to be millions of physical qubits, because a lot of compute on error correction to get a little compute done. And then, about 2030-ish, you’ll start to get to around 1,000 error-corrected qubits, and then our lives will be changed significantly by quantum computers, because at that time you’ll be able to do really sophisticated reverse engineering of material sciences, and later as you get more in the 2,000 to 3,000 errors-corrupted qubit, then what the government worries about is you can break cryptography with a quantum computer, and that’s why you to start doing post-quantum cryptography now, so that anything being transferred over the network is encrypted with something the quantum computer can’t do. So I think that’s important to understand.
Terry Gerton I’m speaking with Bill Vass. He’s Booz Allen’s chief technology officer. Bill, what you just said sort of broke my brain a little bit, but when I’m thinking about the speed of AI and how hard of a time we’re having adapting to that, it sounds like quantum will be even a bigger shift. I wonder, back to sort of the conventional consulting space, how do we build a workforce for this new environment? How is Booz Allen building your own and how are you advising clients?
Bill Vass Yeah, so we have been hiring, I would say, exquisite PhDs right now, and then also bringing in interns and training them in this space. And Amazon did the same thing when I was doing the quantum services at Amazon as well, and lots of people do. Quantum computing is quite a different way of thinking. AI is sort of going to be ingrained in everything we do. But quantum computing is a completely different way of thinking about computing as an analog system, it’s a different way of programming. You’re basically, like I said, creating molecular circuits in the machine’s memory versus connecting together a bunch of steps in a piece of software. And so it’s very different way. So we are going to require a significant amount of training for people to begin to understand that. I mean, in AI, we trained up about 3,200 people at Booz Allen, early days, in AI. We’ve got a lot of AI experts, we’re literally the top AI deliverer for, or implementer for, the federal government today. And we have a lot experience there. And I think we’re probably one of the leaders in quantum as well. And as building that out, as it becomes more real, like I said, we’re doing a lot today with quantum sensing and we’re doing a lot to get ready for quantum computing as it continues to evolve and make sure we understand that. But it is, I think it is a challenge, it’s a very different way of thinking and there’s a very small pool of people who understand it today, relative to the rest of computing.
Terry Gerton Well, that sort of leads me into my next question, because you’ve called this a critical moment for US quantum leadership. But if the people who understand this are in short supply, what needs to happen either technically or politically to make sure we stay ahead?
Bill Vass Well, I think you need to be investing in the supply chain for the quantum computers, both the materials and the manufacturing, you need to be investing in the supply chain for the individuals. So proper training and understanding, and the universities that are going to, that this is going to come out in, or the research labs it’s going to come out in. There is already a lot of government investment in the research labs and the intelligence agencies and a number of other places in quantum. In fact, like so many things, quantum came out of the government, like integrated circuits did, like GPS, like the internet and all these others. Early investments, when I was the chief, one of the chief information officers at the Pentagon in the 90s, I helped to sponsor funding for the early quantum computers that were being built at that time, which is when I first got more familiar with quantum computing. But there’s a lot, quantum computing is, there’s really two fundamental things that are very hard to understand because they’re based on quantum. The first is called superposition. And so when you’re operating on the qubit, it’s in every point of a sphere at once until you measure it. And the second one is entanglement, which allows you to force the molecule or the circuit to do what you want by entangling them together. And entanglement is one of those things, if anyone tells you they actually understand how it works, it’s kind of a chemical bond between two atomic particles that you create in the machine. We know what it does but no one knows how it works, and the way I would describe that is — I’m a motorhead, I understand everything about cars, all the cylinders or the electronics, and I can drive a car, and my wife can drive just as good as I can, but she doesn’t understand anything about the engine or any of those other things — so we can drive superposition and entanglement without fully understanding how it works. But it is a little bit of the frustrating part for me in quantum computing, because I’m one of those people who like to know how everything works, and we know what it does and how to implement it, but we don’t really understand the fundamentals of some of it.
The post Quantum computing – it’s still confusing, but now it’s mission-critical first appeared on Federal News Network.