PsiQuantum bets on light to build the first useful quantum computer
Curated by the Inblix editorial team
The race to build a genuinely useful quantum computer has a new, uniquely photonic contender. PsiQuantum, a company founded in 2016 by four UK physicists, is ditching the superconducting qubits favored by Google and IBM for individual particles of light. Their plan? A machine housed in a room that James O’Donnell describes as a data center crossed with an ice cream factory, featuring 100 stainless-steel cabinets packed with chips that route photons through a maze of optical switches. The core idea is audacious: precisely measuring where each photon ends up could solve problems that would stall classical computers for millions of years.
The machine, of course, does not exist yet. But in a field saturated with deep-pocketed competitors and even deeper hype, PsiQuantum’s approach stands out for its sheer physical scale and unconventional architecture. The company is targeting a fault-tolerant machine, a level of capability that remains elusive across the entire industry. While others grapple with error-prone qubits that need near-absolute-zero temperatures, PsiQuantum’s bet on silicon photonics leverages existing semiconductor manufacturing processes, a pragmatic choice that could accelerate production if the physics holds up.
That’s a big ‘if.’ The challenge of wrangling thousands of individual photons, accounting for each one’s journey through a silicon labyrinth, is a monumental engineering problem. A single misplaced particle could cascade into computational garbage. The story isn’t just about a new device; it’s about a philosophical split in the quantum world—whether the path to supremacy lies in perfecting artificial atoms or in mastering the fundamental particles of light. The sheer complexity of the proposed facility makes it either a stroke of genius or a breathtakingly expensive science project.
It’s easy to be cynical about another quantum ‘breakthrough’ that’s perpetually ten years away. But PsiQuantum’s willingness to think at data-center scale from day one, rather than incrementally scaling a lab experiment, is a genuinely different signal. The company is forcing a conversation about what a post-classical computing world actually looks like physically, not just mathematically. Whether they succeed may depend less on the cleverness of their physicists and more on the grit of their engineers.
💡 Key Takeaways
- PsiQuantum is pursuing a radically different quantum computing architecture using photons instead of superconducting qubits, betting on light particles measured through silicon photonic chips.
- The company's planned machine would be a data-center-scale facility with 100 cabinets, targeting fault-tolerant computation that could outperform classical computers by millions of years.
- Leveraging existing semiconductor manufacturing for its photonic chips gives PsiQuantum a potential scaling advantage, though the engineering challenge of controlling thousands of photons remains immense.
- Unlike competitors scaling from lab experiments, PsiQuantum's industrial-scale design from the outset forces a practical conversation about the physical reality of a post-classical computing era.
Keep reading: See related articles below for more coverage on this topic.
Get smarter about AI
The sharpest AI news, curated daily. Delivered free to your inbox.