The archive · Space, Robots, Defence · Technical decision · 2016–2025
PsiQuantum Bets on Photonic Qubits in Semiconductor Factories
PsiQuantum broke ground in Chicago on 2025-09-30 to build first million-qubit-class fault-tolerant quantum computer in US.
PsiQuantum
What the business is
Builds fault-tolerant quantum computers on the photonic route: uses standard semiconductor processes to make quantum photonic chips, expands them via fiber interconnects into million-qubit-class machines, and sells them to institutional customers that need computing power.
Starting capital:2025-09-10 disclosed Series E of more than $1 billion (official press release); cumulative commitments from the State of Illinois and local governments of approximately $700 million (2024 $500 million + earlier $200 million, according to Quantum Insider).
How it started
Founded in 2016 in Palo Alto, California, by Jeremy O'Brien, Pete Shadbolt and others. The bet began with a route choice: rather than struggling to improve superconducting-qubit coherence, it bet that photonic qubits could be mass-produced using the semiconductor industry's existing capacity—first small-scale validation, then scaling all the way to a million qubits.
What happened
In 2024-07, it reached agreements with the State of Illinois, Cook County, and Chicago to locate at the Illinois Quantum and Microelectronics Park (IQMP); on 2025-09-10, it announced a Series E of more than $1 billion and disclosed that the Brisbane campus in Australia was being advanced in parallel; on 2025-09-30, it broke ground at the former U.S. Steel South Works site in Chicago—first building the company's largest intermediate-scale test system, to be directly reviewed by a DARPA Quantum Benchmarking Initiative (QBI) expert panel, and then building and deploying the first million-qubit-class fault-tolerant quantum computer in the United States.
No ending yet — it is still running.
Background
PsiQuantum, founded 2016, bets on photonic qubits in silicon chips, mass-produced via semiconductor processes, networked over fiber. It defines problem as manufacturability, not coherence.
In 2024-07, chose Illinois Quantum Park as US site; 2025-09-10 announced $1B+ Series E; 2025-09-30 broke ground. First phase builds intermediate test system for DARPA review, then million-qubit computer.
Illinois committed ~$700M to park. Company must prove photonic scaling easier than superconducting. Groundbreaking is start; press release says 'hard work still lies ahead'.
As of 2026-09-01, machine not built. Phase 1 builds infrastructure and 80,000 sq ft building. Significance: startup shifted quantum computing from lab to factory scale.
What has to be true
- Manufacturability of photonic qubits reuses semiconductor industry, bypassing coherence and control-line problems.
- Milestones are verifiable: groundbreaking, campus, Series E, DARPA review are public.
- Capital is large: over $1B Series E plus $700M public investment raise stakes.
- Risks remain: no million-qubit machine built, test system and DARPA review pending.
What can be applied
Treating quantum computing as a manufacturing problem can bypass competitors' hardest parts, but risk shifts to engineering and funding lasting until results.
Aftermath
As of 2026-09-01, PsiQuantum is building. Chicago IQMP Phase 1 includes an 80,000-square-foot office and R&D building with cryogenic environment. Later phases and DARPA review remain. Machine not live; fundraising and campus construction continue.
Sources
- PsiQuantum Breaks Ground on America's Largest Quantum Computing Project in Chicago
- Groundbreaking of Illinois Quantum and Microelectronics Park creates anchor for quantum innovation
- Chicago Breaks Ground on Quantum Campus at Former South Works Site
spotted an error? The archive wants to know.
Your turn
You just read one. Describe what you are building, and see who is betting on the same thing.
Free account · 3 free questions · no card