Department of Energy launches $215 million race to build quantum computer by 2028

Quantum computing researchers, start your engines! The U.S. Department of Energy (DOE) announced yesterday it will sponsor a $215 million competition in which teams will race to build a scientifically useful, error-corrected quantum computer by 2028. DOE envisions supporting between three and 10 teams, and to get full funding, a team will have to reach various performance milestones along the way. The competition is “a testament to our decades of support and investment in basic science,” says Darío Gil, DOE’s undersecretary for science, who spoke with reporters on a 16 September press call. “There are moments where we can realize incredible technology and capability as a consequence of that investment, and we are in one of those moments.” A quantum computer would exploit the strange rules of quantum mechanics to solve certain computational problems that would overwhelm any conventional supercomputer, such as cracking certain encryption schemes and deciphering the structures of molecules. It would gain its advantage by encoding many potential solutions to a problem as different quantum waves sloshing simultaneously through so-called qubits, the quantum counterparts of the microscopic switches, or bits, in conventional computers. Researchers have fashioned qubits out of individual atoms suspended in laser light, ions trapped on a microchip, tiny resonating circuits made of superconducting metal, and other technologies. Set the states of these qubits just right and waves representing wrong solutions will cancel one another while the wave representing the right one stands out. However, the states of qubits are fragile and prone to disruption from the environment. To prevent such muddling, researchers spread the state of a single “logical” qubit over many physical qubits in schemes that enable the logical qubit to hold its state much longer than any of the physical qubits. A quantum computer would be fault tolerant if such an error correction scheme suppressed logical qubits’ errors to a negligible level. Federal agencies and private companies have already poured billions of dollars into quantum computing and have developed machines comprising a few hundred qubits. But technologies have advanced enough that much bigger and more capable machines are possible, Gil says. DOE envisions a fault-tolerant machine with between 100 and 200 logical qubits, which would likely require thousands of physical qubits. Building such an unprecedented machine by 2028 will be a challenge, Gil acknowledges. “Do I think it’s possible?” he says. “Yes, absolutely. Am I certain it will happen? No.” Tanner Crowder, quantum information science lead and senior technical adviser for advanced scientific computing research in DOE’s Office of Science, says each winning team will have to explain how it can validate and verify the quantum performance of its machine, present a prototype, and scale up to perform a scientifically useful computation. “Our goal is not to have a bunch of qubits that can’t do much,” Crowder says. “We want to be able to see meaningful scientific calculations done on these quantum computers.” DOE will soon issue a funding opportunity announcement and hopes to select teams by December, Crowder says. “We hope to start work as soon as possible,” he says. “There’s an obvious urgency to us being able to deliver this capability for the nation.”