Nvidia in advanced talks to invest in quantum startup PsiQuantum, reversing Jensen's earlier skepticism
May 19, 2025
Key Points
- Nvidia is in advanced talks to invest in PsiQuantum, marking its first bet on physical quantum hardware and reversing CEO Jensen Huang's January skepticism about the field's near-term viability.
- Huang shifted strategy within months, hosting Nvidia's first Quantum Day in March and opening a Boston quantum research center, signaling a hedge against what some see as a computing wave bigger than AI.
- PsiQuantum, valued at $3 billion after raising over $1 billion from investors including BlackRock and Microsoft's M12, is racing to build quantum computers for drug discovery despite the field remaining largely theoretical and production-scale deployment years away.
Summary
Nvidia is in advanced talks to invest in PsiQuantum, a Palo Alto-based quantum computing startup. The move reverses Jensen Huang's skepticism from January, when he called quantum computing likely two decades away from usefulness. By March, Huang hosted the first Quantum Day at Nvidia's developer conference and announced a new quantum research center in Boston.
The investment would be Nvidia's first bet on a company building physical quantum hardware. Nvidia previously invested in Sandbox AQ, which raised over $450 million to develop quantum software algorithms. PsiQuantum has raised more than $1 billion from investors including BlackRock, Playground Global, Microsoft's venture arm M12, and the Australian government. The startup was valued at $3 billion after a 2021 fundraise of $450 million.
Unlike most Nvidia investments, which target startups that buy or rent large amounts of its GPUs, the PsiQuantum deal signals a different strategy. Nvidia is positioning itself ahead of what some believe could be a computing wave bigger than AI, though one still years away. PsiQuantum is racing to build quantum computers capable of outperforming traditional systems on tasks like simulating molecular interactions for drug discovery.
Quantum computing remains largely theoretical. No major company has shipped meaningful production technology. The field spent decades in academia before IBM and Google began public testing. Rigetti Computing, a quantum startup that went public via SPAC, gained significant valuation and stock appreciation despite not delivering production-ready systems at scale. The sector appears positioned similarly now, with scientific progress and strong teams commanding high valuations even without real-world deployment.
Huang's pivot shows how quickly strategic positioning can shift. The quantum research center and expanded conference presence suggest Nvidia is hedging against a long-term technology bet while simultaneously extracting near-term value from its GPU infrastructure positioning.