PsiQuantum’s $1B Series E, Nvidia, & the Race to 1 Million Qubits
Timestamps are as accurate as they can be but may be slightly off. We encourage you to listen to the full context.
Podcast Summary
PsiQuantum CEO Pete Shadbolt joins Sorcery to discuss the company's recent $2 billion funding milestone and their ambitious vision to build the world's first commercially useful quantum computer. (00:00) The company is taking a contrarian approach by skipping incremental development and going directly to million-qubit systems - a scale needed for true commercial impact. (06:24)
Main Theme:
- PsiQuantum's unique strategy of bypassing small-scale quantum demonstrations to build utility-scale quantum computers that can actually solve commercially valuable problems in chemistry, materials science, and drug discovery
Speakers
Pete Shadbolt
Pete Shadbolt is the Co-founder and CEO of PsiQuantum, one of the world's most well-funded quantum computing companies with nearly $2 billion in private capital raised. A British physicist who moved to Silicon Valley a decade ago, Shadbolt previously conducted foundational research in optical quantum computing at the University of Bristol and Imperial College. He put one of the first quantum processors online in the UK more than a decade ago and has been instrumental in establishing PsiQuantum's partnerships with major semiconductor manufacturers like GlobalFoundries.
Molly Mielitz
Molly Mielitz is the host of Sorcery, a podcast focused on frontier technology and deep tech investments. She conducts in-depth interviews with leading entrepreneurs and investors in emerging technology sectors, with particular expertise in quantum computing, AI, and advanced manufacturing.
Key Takeaways
Skip the Incremental Approach for Breakthrough Technologies
Shadbolt emphasizes that PsiQuantum deliberately rejected the conventional startup path of building incremental improvements. (07:00) Instead of creating small quantum systems for demonstrations, they committed entirely to developing million-qubit machines from day one. This approach required hundreds of millions in upfront investment but positions them to deliver actual commercial utility rather than impressive but useless demonstrations. The analogy Shadbolt uses is powerful: if you need to get to the moon, building taller and taller ladders isn't the right approach - you need to invest in rocket engine technology.
Leverage Existing Infrastructure for Exponential Scaling
Rather than building everything from scratch, PsiQuantum strategically integrated with existing semiconductor manufacturing infrastructure. (14:52) By designing their quantum technology to work within established fabs like GlobalFoundries and existing supply chains, they can achieve the kind of rapid scaling that Jensen Huang demonstrated with the xAI Colossus supercomputer - 100,000 GPUs in 112 days. This approach allows them to benefit from the trillion dollars and fifty years humanity has invested in semiconductor infrastructure.
Focus on Problems Where You Have Categorical Advantage
Quantum computers excel at simulating quantum systems - the microscopic foundations of chemistry, materials science, and drug discovery that conventional supercomputers fundamentally cannot handle well. (23:54) Shadbolt describes these applications as "virgin territory that will otherwise remain forever unexplored" by classical computing. Rather than competing with AI in areas like language processing, quantum computing should focus on the physical world problems where it offers an insurmountable advantage.
Build for the Frontier, Not the Market
The most valuable companies of our era - TSMC, ASML, SpaceX, NVIDIA - are all "living or dying on the quality of their science" and operating "way out on the frontier." (21:54) These companies capture enormous value precisely because they do something so technically difficult that nobody else can replicate it. Power is captured on the frontier of advanced technology, not in incremental improvements to existing solutions.
Embrace Zero-to-One Technology Development
Shadbolt argues against the common venture capital advice to generate early revenue "by any means necessary." (50:48) For truly revolutionary technologies like quantum computing, attempting to create incremental revenue streams with fundamentally inadequate technology creates false metrics and diverts resources. Like Waymo with autonomous driving, quantum computing companies should accept being "useless until they cross the threshold" to actual utility.
Validate Through Red Team Adversarial Testing
The most credible validation comes from skeptical third parties actively trying to disprove your approach. (42:24) DARPA's red team exercise, where 50 scientists and engineers were tasked with essentially "killing quantum computing companies," provided more meaningful validation than typical due diligence. After years of scrutiny, PsiQuantum was one of only two companies to graduate to the final phase, demonstrating the robustness of their technical approach.
Plan for Vertical Integration from Discovery to Application
While initially planning to sell time on their quantum computer like cloud computing services, Shadbolt envisions vertical integration - hiring drug discovery teams, material scientists, and chemists to directly exploit the knowledge generated by their quantum systems. (55:37) This approach could create "a company the likes of which the world has never seen before" by controlling the entire value chain from quantum simulation to commercial application.
Statistics & Facts
- PsiQuantum has raised nearly $2 billion in private capital, more than any other quantum computing company, including a recent billion-dollar deal with the Australian government. (05:25)
- The general consensus is that you need approximately one million qubits to do anything commercially impactful with a quantum computer, while Google's recent systems demonstrate around 100 qubits. (06:33)
- PsiQuantum has spent over $100 million over eight years to establish their manufacturing process at GlobalFoundries, with 80% of their supply chain based in the United States involving hundreds of companies. (92:07)
