Quantum computing startup PsiQuantum raises $1B, aiming to build 1M-qubit-scale machine – News

PsiQuantum Inc., a well-funded quantum computing startup, said today it has raised $1 billion in late-stage funding to accelerate its efforts to build a large-scale, reliable quantum computer with more than a million qubits.

The Series E deal was led by BlackRock Inc., alongside Temasek and Baillie Gifford. The fundraise brings the valuation of the company to $7 billion, more than doubling the company’s valuation since its $450 million Series D in 2021. The round brought on more than a half-dozen new investors, including Macquarie Capital, Ribbit Capital, Nvidia Corp.’s venture arm NVentures andAdage Capital Management, with participation from existing investors that include Blackbird, Third Point Ventures and T. Rowe Price Associates Inc.

Quantum computers at any scale require error correction and at a scale as large as 1 million qubits, doing so will become a challenging proposition. Qubits are the quantum equivalent of bits in classical computers. Qubits are used to perform calculations similar to traditional computing; however, they possess properties that allow them to represent both a 1 and a 0 simultaneously. This allows quantum computers to perform extremely complex calculations faster than the largest supercomputers currently available.

“Only building the real thing — million-qubit-scale, fault-tolerant machines — will unlock the promise of quantum computing,” said Prof. Jeremy O’Brien, PsiQuantum co-founder and chief executive. “We defined what it takes from day one: this is a grand engineering challenge, not a science experiment.”

To meet practical demands, quantum computing firms have been pushing machines to larger and larger sizes. Although only suited for very specific computations, quantum computers have shown promise for solving problems in drug discovery, genetics and materials science. When combined with supercomputers in hybrid systems, quantum computers can solve problems in minutes that would take a traditional system months or years.

Since 2021, PsiQuantum has scaled its manufacturing process for its integrated photonic chipset, which uses light for storage and communications within the quantum computer. Additionally, the company has incorporated Barium Titanate into its manufacturing process; this is ideal for ultra-high-performance optical switches, which the company states are the missing component for scaling its machines.

The company said the funds will help it build its ambitious quantum machine by 2028.

PsiQuantum is not the only quantum outfit with plans to build large-scale machines before the end of the decade. IBM Corp. revealed a roadmap with plans for IBM Quantum Starling, a quantum computer capable of performing 20,000 times more operations than today’s machines.

The investment comes shortly after rival Quantinuum raised $600 million at a $10 billion valuation this month and Google LLC unveiled Willow, its largest error-correcting quantum computing chip, at the end of 2024.

Along with the funding, the company announced a collaboration with Nvidia that will cover the development of quantum algorithms and software, and the integration of graphics processing units with quantum computing chips using its infrastructure. GPUs are a fundamental technology used for the training, deployment and computational acceleration of artificial intelligence models.

“AI is built on classical computing, which has underpinned the last fifty years of technology,” said Tony Kim, head of the Fundamental Equities Technology Group at BlackRock. “Now, we are at the dawn of an adjacent computing platform — rooted in quantum mechanics — that will allow us to simulate the physical world with transformative accuracy.”

PsiQuantum also develops the cooling, networking and control systems for utility-scale quantum computers. By harnessing light for networking and control, the company said its systems don’t depend on the now iconic-looking “chandelier”-style cryostats often used to chill quantum computers. Using its specialized technology, the company designed a new high-density cooling solution that can be slotted into modular racks at data centers, capable of cooling hundreds of chips in a single cabinet.

Image: News/Microsoft Designer