We’re excited to announce our investment in PsiQuantum, as part of the USD $450M Series D round.
PsiQuantum is building the first useful quantum computer, with a silicon photonics approach that uses light and well-understood optical components to encode and manipulate units of quantum information (referred to as qubits).
“Nature isn’t classical, dammit, and if you want to make a simulation of nature, you’d better make it quantum mechanical, and by golly it’s a wonderful problem, because it doesn't look so easy.”
Quantum computing has the potential to be the most profound enabling technology of our lifetime, providing us with the tooling required to truly understand chemistry and biology for the first time.
While remarkable progress has been made in classical computing architectures since the microprocessor was released five decades ago, the decline of Moore’s Law and the end of Dennard Scaling mean that the calculations required to accurately simulate the building blocks of life remain beyond our reach without a radically new approach.
The building blocks of life (subatomic particles, like electrons) are quantum mechanical systems, and their behaviour determines the properties of atoms, molecules and compounds that we interact with each day. In order to understand natural processes in chemistry and biology, and predict how the new processes that we create will behave, we need a compute modality that natively speaks the language of life.
It may be hard to believe, but we still lack the tools to accurately simulate the most important processes in our physical world. We do not truly understand photosynthesis or biological nitrogen fixation - despite the fact that these two processes sustain much of life on Earth. How do plants so elegantly store energy in carbohydrate bonds? How does Nitrogenenase catalyse the conversion of atmospheric nitrogen and hydrogen into ammonia at room temperature while we produce ammonia industrially using a resource-intensive process, the Haber-Bosch process, that consumes 2% of global energy production and emits 400Mt of carbon dioxide into the atmosphere each year?
Enter quantum computers, which can encode information in quantum mechanical states and leverage the spooky properties of superposition and entanglement for information processing.
We can think of quantum computers as enabling massive parallel computation and the ability to perform a series of complex calculations that will remain intractable on classical computers. It would take many more bits than there are atoms on Earth to perform these same calculations on a classical computer.
Quantum computers will allow us to model the building blocks of life with high fidelity, and unlock the knowledge that we so desperately need to advance the grand challenges of our generation.
We hope that we can use this technology to help fight climate change by designing more efficient battery chemistries, photovoltaic cells, and catalysts for carbon capture.
We are optimistic that we might use quantum computing to develop new in-silico techniques for screening candidates, and delivery systems to efficiently bring to market the next generation of therapeutics that will treat currently incurable diseases.
It is hard to build a quantum computer, and much harder to build one that is #useful# in helping to solve our most painful problems.
The quantum mechanical states used in computation are fragile, and the sophisticated components that are used to create and manipulate these states are delicate. This means that a useful quantum computer needs an effective error correction scheme to ensure that the output remains reliable. It is broadly accepted that useful quantum computers will need to be fault tolerant (resistant to errors) and support at least one hundred error-corrected qubits (referred to as logical qubits).
This is a considerable task. Most error correction schemes are redundancy based and require on the order of one million “raw” qubits in order to run commercially valuable applications. There is a 10,000x difference between the current state-of-the-art in quantum hardware and this one million raw qubit target.
The real challenge on the path to useful quantum computing is not to demonstrate that we can produce qubits (we know that we can) but to demonstrate that we can produce and precisely control enough raw qubits to implement fault-tolerant quantum computation at scale.
PsiQuantum believes that the path to a useful quantum computer will not be achieved by incrementally increasing the number of raw qubits. Instead, it has designed a high-conviction technical roadmap that is focused on overcoming all of the challenges that lay on the critical path to a million qubit machine, upfront.
Scalability is an important advantage that silicon photonics has over other approaches to quantum computing, and PsiQuantum’s architecture leans into this advantage.
The magic is that PsiQuantum is taking an existing scalable process and working to make it quantum, rather than taking a novel quantum process and working to make it scale.
Its architecture does not require exotic materials or fabrication processes, which means that PsiQuantum can access the mature, high-performance 300mm production lines at tier-one semiconductor foundries. PsiQuantum has the sophistication of more than five decades and trillions of dollars invested in semiconductor fabrication behind it.
PsiQuantum recently announced its manufacturing partnership with GlobalFoundries and has already demonstrated the ability to produce high-quality single-photon sources and single-photon detectors at scale using standard manufacturing processes.
There are still significant technical challenges to overcome on the path to building the first useful quantum computer, but PsiQuantum has the most advanced tooling and the most experienced semiconductor engineers at its disposal.
If PsiQuantum can overcome these challenges, its architecture can be arbitrarily scalable.
It is hard to imagine a more compelling example of founders pursuing their life's work, or a founding team that is better equipped to overcome the challenges that lay ahead.
PsiQuantum is led by a team of four technical founders (Jeremy O’Brien, Pete Shadbolt, Mark Thompson, and Terry Rudolph), including two Australians, who are recognised as global leaders in photonics quantum computing, theoretical quantum physics and ultrafast optical components. The founding team were collaborators for more than a decade in academia at the University of Bristol and Imperial College London before establishing PsiQuantum in 2017.
To date, quantum computing has promised a lot but delivered very little. PsiQuantum will change this. It will change us. The first useful quantum computer will prompt us to reimagine our relationship with the natural world, and our role as its custodians. The vision that PsiQuantum presents is truly science non-fiction; the most consequential enabling technology of our lifetime.
We are investing a little later than usual here but that should tell you exactly how excited we are about the opportunity to partner with Jeremy, Pete, Mark, Terry, and their team at PsiQuantum.