DARPA plans ‘major expansion’ of quantum computing push next month

A graphic concept of quantum technology depicts two ships at sea in the background and a large, stylized molecule in the foreground. (U.S. Navy illustration by NIWC Pacific/Released.)

WASHINGTON — The self-proclaimed “quantum skeptic” heading DARPA’s Quantum Benchmarking Initiative (QBI) tells Breaking Defense he is “pleasantly surprised” by how many companies’ quantum computing plans look feasible — and as a result, he’s weeks away from announcing a slew of new awards.

The point of the QBI program is not to pay the private sector to develop the tech, which they’re already doing on their own, according to program manager Joe Altepeter. Instead, in essence, a company that wins a QBI award gets access to DARPA’s interagency team of expert testers and devil’s advocates, serving as an independent outside check to make sure that they’re on a practical path.

Tech titan Microsoft and specialist startup PsiQuantum have already won contracts for the initial, pilot-project version of the program, known clunkily as Underexplored Systems for Utility-Scale Quantum Computing or US2QC. (“You can tell why we changed it to ‘QBI,’” Altepeter said). Two other companies, Atom Computing and QC82, also won preliminary, short-term grants in 2023, but their proposals didn’t survive subsequent rounds of more intensive scrutiny. So only two of four competing firms made it all the way through the three-phase process to win one of the long-term “Stage C” contracts announced Feb. 6.

That 50 percent success rate is actually much better than Altepeter expected.

“That was definitely surprising,” he told Breaking Defense. “I didn’t think anybody was going to make it to Stage C. But that’s why we do this, to get surprised.”

So a “major expansion” to “bring in many more companies” is now underway, Altepeter said. “We’ve gotten finished with source selection [and] sent out letters inviting all the companies who’ve made it into stage A to negotiate a contract…. My best guess is late March, we’re going to have the vast majority of those contracts signed.”

RELATED: Pentagon info officer’s top priority: Upgrading cryptography ahead of quantum-enabled hackers

In this new round of competition, once again, Altepeter said many competitors exceeded his expectations for how far along their tech is and how feasible their plans are to build reliable, practical, and usefully powerful quantum computers. As a result, “we’ll have the chance to kick the tires on a lot more companies,” he said. “A year from now, we’ll know a lot more about if this industry is for real or not.”

How does that timeline work? The process is meant to be flexible, with the schedule accommodating what companies think is required, and companies that fail at one stage of the evaluation — like Atom and QC82 — are welcome to come back for a second shot. But overall, Altepeter said, QBI goes like this:

• First, interested companies submit written proposals and go through four hours of oral presentations with a board of experts.
• Companies that pass this review get a Stage A contract for a six month “sprint.” This is when they lay out their overall vision for their future quantum computer and try to convince DARPA it will really work at tasks that really matter. As Altepeter said, “What’s it going to do? Why do you think you can get there? Why is it important?”
• Companies that pass Stage A move on to a more intensive 12-month assessment, Stage B, in which they lay out their entire quantum R&D plan for DARPA to look at dozens of aspects, “digging deep” into everything from technical conundrums like quantum error correction to the economics of their business case.
• Finally, companies that survive Stage B get a contract to enter the long-term, open-ended Stage C. “There’s no set time” for Stage C, Altepeter explained. “It’s whatever it’s going to take according to their R&D plan [and DARPA will] follow along. Let’s do independent tests. Let’s have people looking over your shoulder in your labs. Let’s have some of your components being tested in our labs.”

At each stage, the size and cost of the DARPA team roughly doubles, Altepeter said. Currently, it consists of roughly 200 quantum experts from across the government — including Energy Department labs like Los Alamos as well as Defense Department labs like the Air Force Research Laboratory — and government-funded research centers, such as Johns Hopkins. (These personnel will remain with their home organization but spend roughly 60 to 70 percent of their time on DARPA quantum work). The idea, he said, is to get independent, expert feedback from professionals with no financial stake in the outcome either way.

With new technologies, it’s all too easy to get sucked in by hype cycles and impressive statistics, Altepeter said, so it’s critical, and difficult, to figure out the right way to measure progress. Track the wrong metrics, he warned, and you end up with the equivalent of assessing a space-launch program by highest altitude achieved to date, a superficially reasonable approach that ends up grading grifters who can climb a ladder over professionals conducting static rocket-motor tests.

“We are building by far the best, largest test and evaluation team for quantum computing on the planet,” Altepeter summed up. “[We’re] planning for this to be a $1 billion-plus program.”

This kind of careful evaluation isn’t cheap, he acknowledged, and if too many companies prove qualified, he may not have enough funding to give all of them awards.

“It’s a double-edged sword,” Altepeter said. “I’m pleasantly surprised that these companies were doing so well. I also worry that we couldn’t eliminate as many companies, because we’re definitely trying to eliminate companies.”

Could the Trump Administration’s controversial ongoing cuts to government agencies, spearheaded by Elon Musk’s DOGE, cramp or cripple the Quantum Benchmarking Initiative just as it’s trying to grow? Altepeter demurred on those “bigger-scope…budget issues.”

That said, when it comes to federal funding, “it’s always a danger that there are other priorities that take precedence,” he said. “But my marching orders are the same as they have been since we started this…which is to sprint as fast as possible and get ready for a world-changing program.”