DARPA taps OTAs to accelerate the making of a useful quantum supercomputer
An ultramodern system might not be so far off.
U.S. defense researchers recently moved to partner with the private sector to strategically explore building the world’s first practical quantum supercomputer.
“There's a lot of hype in the commercial space and there's a lot of people claiming that they've figured out a path to a really big, really useful quantum computer. And we would like to listen—like, if somebody thinks that they cracked the secret code to make any quantum computer, then we would love for them to apply for this program,” Joe Altepeter, a program manager in the Defense Advanced Research Project Agency’s Defense Sciences Office told Nextgov this week. “And we want to be really flexible in how we work with companies.”
Altepeter is a subject matter expert at the key national security research arm, where, among other programs, he’s leading the newly unveiled Underexplored Systems for Utility-Scale Quantum Computing or US2QC effort. Through it, researchers intend to puzzle out whether any approach is capable of achieving a utility-scale operational quantum machine quicker than conventional predictions.
In a recent conversation, Altepeter briefed Nextgov on DARPA’s vision for the new program, what it might enable, and the complexities that render quantum pursuits so intensely complicated.
“That's really what we do here. Is there a crazy idea that nobody else is going to pursue that has a chance of making the world a better place? That's what the DSO is all about,” he said.
A ‘DARPA-hard’ problem
Quantum computing is complex and emerging. It involves employing the bizarre phenomena of quantum mechanics to deliver what might be game-changing advances in computation.
But for now, the operative word here is “might.”
“When I think about quantum computing, in particular, I think about, say, the 10 smartest physicists that I've worked with in my life. About half think that it's going to be the most important technology of the twenty-first century—it's going to revolutionize everything, bring huge boosts to our [gross domestic product], it's fantastic. And the other half think that even if you could build a quantum computer, which you cannot, it wouldn't do anything better than a classical supercomputer,” Altepeter explained. “And when it’s that chasm between like brilliant people on both sides—and some think it's going to be the best thing since sliced bread, and some think that is just going to be useless—that just screams that DARPA needs to step in, because fundamentally our mission is to prevent strategic surprise.”
US2QC marks the latest quantum-driving effort that research hub is embarking on.
Other federal components beyond DARPA are also investing in and pursuing programs to develop quantum systems and explore associated use cases. That’s partly because of all the possibilities it represents. Some predict that quantum advancements could eventually lead to almost unimaginable breakthroughs in medicine or national security, like enabling unhackable communications.
“People think it might be everything, and some people think it might be nothing. And we realize that, to really answer this—it's not an easy answer,” Altepeter noted. “But it's actually a really difficult problem. It’s what we call a DARPA-hard problem to try to figure out which paths to revolutionizing something are really applicable.”
During the interview, he shed light on some of the intricacies that make quantum computing development so complicated.
“One of the reasons we're hopeful that quantum computers will be so powerful is, in a sense, they have a lot more information than a classical computer,” Altepeter explained.
Classical computers are made of basic units called bits, and experts generally have a strong idea of of how many they’re comprised of. Quantum computers are based on quantum bits, or qubits that can exist in multiple states at one given time. While a classical machine’s bit represents a one or zero, quibits could be both of those, or what’s in between. Altepeter noted that what’s going on inside quantum computers is “literally exponentially bigger” than digital computers.
“And I don't mean in the way people normally use ‘exponentially’ to mean ‘a lot bigger.’ I mean, you know, after you get to a computer that's 100 quantum bits, it's got more degrees of freedom than there are atoms on Earth. I mean, it's like ludicrously huge,” he said. “And so there's a hope that that kind of bigger space it plays in gives it power. But try to keep track of how it gets there—like there's just so many ways it can wander off course, and be doing something that you didn't intend it to do.”
Another “romantic” idea about quantum mechanics that draws people into this fresh realm of computing and captures the imagination is, as Altepeter put it, “when you measure something in a quantum system, you change it—and in a sense, you become entangled with that system.”
In practice, this means that any atom or even piece of light that “talks” to the quantum system, can change the computation that's going on, unless it’s built with exquisitely complex mathematics that enable it to be fault tolerant.
Some foundations of this concept were completed by mathematicians as opposed to physicists.
“It’s still really an open question if that kind of mathematics can be translated into engineering. And so using a much bigger, much more complex system to implement this beautiful multi-dimensional mathematical object that protects your data from everyday atoms jostling it is what we would need. Exactly how we do that, and if it's possible to work—nobody knows yet. Hopefully, the US2QC program helps to answer that question.”
DARPA released a detailed program solicitation for entities and organizations interested in creating a truly useful, fault-tolerant quantum computer. They’re invited to submit initial abstracts by March 23.
“We want to go into this with our eyes open, knowing that there's a lot of companies who are already doing this,” Altepeter noted.
The research arm’s team wants to engage with those who already have a credible plan and path to a useful quantum computer—and potentially help fund that work.
“We'd love to give them a source of unbiased verification and validation, which is something that the government's really in a kind of unique position to give. And we'd like to be able to effectively communicate—as kind of a neutral third party to other stakeholders in the government—what the real situation on the ground is,” Altepeter said. “But that really calls for a very flexible approach, right? People are not going to want to be involved—and rightly so—in a program that would slow them down or change their trajectory to a solution that we're prescribing.”
There are heaps of ways to build a quantum computer, so the US2QC is structured for maximum flexibility.
To fund the proposals, DARPA will exclusively use tailorable other transaction agreements. The only common foundation for all proposals is known as Phase 0. For it, proposers will describe a complete utility-scale concept for the futuristic system, and then only some will move forward.
Though it is not guaranteed that a useful quantum machine will be built in this process, Altepeter said DARPA is “always open to being surprised” in its research.
“I never want to say that there's a limit to how much a DSO program can accomplish,” he noted.
This work is unfolding as the Defense Department also faces a fresh and serious conflict abroad sparked by Russia’s ongoing invasion of Ukraine. While that requires new focus and a sense of urgency from DOD, Altepeter said it’s critical that DARPA continues to hone in on the conflicts and challenges that will happen years from now.
“If we ever decide that we're going to stop looking to the future and figuring out how to solve the problems of the next few decades, I think that just means that in the next few decades we're going to be in a really sorry state,” he said.