China vs USA: The Quantum Computing Race

In October 2019, Google made headlines around the world. Its 53-qubit Sycamore processor had performed a calculation in 200 seconds that, the company claimed, would take the world's fastest supercomputer 10,000 years. Google called it "quantum supremacy" — the moment a quantum computer first did something no classical computer could practically do. The milestone was contested — IBM argued the supercomputer could do it in 2.5 days with better algorithms — but the symbolism was clear: the United States had fired the first shot in the quantum computing arms race.

Six years later, the race looks very different. In 2025, a team led by Pan Jianwei at the University of Science and Technology of China (USTC) announced that their Zuchongzhi 3.0 processor (祖冲之三号), with 105 qubits, had solved a sampling problem roughly 10 million times faster than Google's latest result — a calculation that would take the world's most powerful classical supercomputer approximately 6.4 billion years. A separate Chinese team using the Jiuzhang 3.0 (九章三号) photonic quantum computer claimed a comparable breakthrough using a completely different physical approach. China had not just matched the United States in quantum computing. On at least one important benchmark, it had pulled ahead. The question now is whether that lead is durable — and what it means.

Two Roads to Quantum Supremacy

To understand the quantum race, you first have to understand that there are multiple fundamentally different ways to build a quantum computer. The current front-runners are superconducting qubits and photonic qubits — and China is competitive in both.

Superconducting qubits — tiny circuits made of superconducting metal that behave according to quantum mechanical rules at near-absolute-zero temperatures — are the approach used by Google, IBM, and the Chinese team behind Zuchongzhi. They are the most mature quantum computing technology, but they require elaborate cryogenic cooling systems that keep the processor at roughly -273 degrees Celsius (colder than deep space). Every additional qubit makes the system exponentially harder to cool and control. The world's best superconducting processors are in the 100-1,000 qubit range, with error rates that still make them unsuitable for most practical applications.

Photonic quantum computing — using particles of light (photons) instead of superconducting circuits — is the approach behind China's Jiuzhang. It has one enormous advantage: it operates at room temperature. No cryogenic cooling. No billion-dollar refrigeration infrastructure. But it has a corresponding disadvantage: photonic qubits are much harder to control precisely, and scaling up the number of photons while maintaining quantum coherence is an unsolved engineering challenge. Jiuzhang 3.0 uses roughly 255 photons, but the system is not programmable in the way Google's Sycamore is — it can only solve a specific class of problems called boson sampling.

This is the key caveat behind every quantum supremacy claim to date. The problems that Google and China's teams have solved — random circuit sampling and boson sampling — are synthetic benchmarks with no known practical application. They prove that quantum computers can in principle outperform classical computers on some tasks. They do not prove that quantum computers can do anything useful. The transition from "quantum supremacy" to "quantum advantage" — where a quantum computer does something commercially valuable that a classical computer cannot — has not yet happened for anyone, Chinese or American.

Where the US Leads: The Ecosystem

If quantum computing were just about who has the most qubits, China might be winning. But quantum computing is not just about qubits. It is about an entire ecosystem — and here, the United States remains firmly ahead.

First, there is the software stack. IBM has built Qiskit, an open-source quantum computing SDK (software development kit) that has become the de facto standard for quantum programming. Google has Cirq. Microsoft has Q# and Azure Quantum. Amazon has Braket. These platforms allow researchers and developers to write quantum algorithms, simulate them on classical hardware, and run them on real quantum processors in the cloud. China has equivalents — OriginQ's QPanda, Baidu's Paddle Quantum — but they are years behind in terms of developer adoption, documentation quality, and third-party library support.

Second, there is the talent pipeline. The United States graduates roughly three times as many PhDs in quantum information science as China, and its top institutions — MIT, Caltech, Stanford, the University of Chicago — have built quantum research centers that attract the best students from around the world. China is closing the gap: USTC has arguably the best quantum research program in the world outside the United States, and Chinese PhD production in quantum science is growing at roughly 15% per year. But the absolute numbers still favor the US, and the US advantage is amplified by its ability to attract international talent — including, critically, from China itself. Many of the top Chinese-born quantum researchers in US labs have chosen to stay in the United States.

Third, there is the venture capital ecosystem. American quantum startups — IonQ, Rigetti, PsiQuantum, Quantinuum — have raised billions of dollars from private investors and gone public via SPACs (special purpose acquisition companies). Chinese quantum startups — Origin Quantum, SpinQ, CIQTEK — have raised significant funding, but at roughly one-tenth the scale. The US quantum venture ecosystem is deeper, more liquid, and better connected to the global market. Quantum computing is a capital-intensive field. Having more capital matters.

Where China Leads: State Commitment and Speed

China's quantum advantage is not in startups or software. It is in state commitment — and that commitment is enormous.

China's 14th Five-Year Plan (2021-2025) identified quantum information as a top national priority, alongside AI, semiconductors, and brain science. The government has committed an estimated $15 billion to quantum research through 2030 — roughly three times the US federal investment when adjusted for purchasing power parity. The National Laboratory for Quantum Information Sciences, headquartered in Hefei, is one of the largest quantum research facilities in the world, with over 1,000 researchers. China has also built the world's longest quantum communication network — a 2,000-kilometer fiber-optic link between Beijing and Shanghai — and launched the world's first quantum communication satellite, Micius, in 2016.

This state-led approach has one decisive advantage: it can fund things that venture capitalists would never fund. The Jiuzhang photonic quantum computer, for example, is a scientific instrument with no near-term commercial application. No venture capitalist would invest in it. But the Chinese state did, because it views quantum computing as a strategic technology, not a commercial one. The same is true of China's quantum communication network — a technology that has no obvious business model but significant military and intelligence applications. In the United States, quantum communication research is almost non-existent because there is no market for it. In China, it is a major national project.

China's speed advantage is also real. The Zuchongzhi processor went from 66 qubits to 105 qubits in roughly two years — a pace of improvement that exceeds any American quantum lab. Chinese quantum research papers now account for roughly 30% of the global total, up from under 10% a decade ago. On patent filings, China leads the world by a wide margin, with roughly twice as many quantum-related patents as the United States. Patent counts do not equal quality, but they do equal intent. China intends to lead in quantum computing, and it is spending like it.

The Encryption Overhang

The quantum race matters for reasons that go beyond computing speed. It matters because a sufficiently powerful quantum computer could break most of the encryption that protects the internet.

Modern internet security relies on mathematical problems that are easy to verify but extremely hard to solve — factoring large numbers (which underlies RSA encryption) and computing discrete logarithms (which underlies elliptic curve cryptography). A classical computer would take billions of years to crack a 2048-bit RSA key. A sufficiently powerful quantum computer running Shor's algorithm could do it in hours. This is not theoretical. It is a proven mathematical fact. We just do not have a quantum computer powerful enough to run the algorithm yet.

The National Institute of Standards and Technology (NIST) in the United States has been working on post-quantum cryptography — encryption algorithms that even a quantum computer cannot break — and finalized its first standards in 2024. The US government has mandated that all federal systems migrate to post-quantum encryption by 2035. China has launched a parallel effort through the State Cryptography Administration, with its own post-quantum algorithms and its own migration timeline. The two countries are not coordinating. They are racing — and the winner will determine whose encryption standards become the global norm.

The nightmare scenario, for either side, is that the other achieves a cryptographically relevant quantum computer first — a machine powerful enough to break existing encryption — before post-quantum defenses are widely deployed. Every encrypted message sent today can be stored and decrypted later. If a quantum computer capable of running Shor's algorithm emerges in 2035, it will retroactively break the encryption on every intercepted message sent before 2035. This is why intelligence agencies on both sides are investing heavily in quantum computing. The quantum race is not just about who can compute faster. It is about who can read the other's secrets.

The Verdict

The quantum computing race between China and the United States is closer than most people realize, but it is not yet a race that either side is clearly winning. The United States leads on the ecosystem — software, talent, and venture capital — and its companies are better positioned to commercialize quantum computing if and when it becomes commercially viable. China leads on state investment, experimental results, and deployment speed — and its photonic and superconducting processors have produced headline-grabbing benchmark results.

But benchmarks are not products. The trillion-dollar question is who will be the first to achieve "quantum advantage" — a quantum computer that does something economically useful that a classical computer cannot. No one has done it yet. Google's 2019 result did not. China's 2025 results did not. The consensus among physicists is that fault-tolerant, commercially useful quantum computing is still 10-15 years away — and that the winner of the quantum race will be determined not by who has the most qubits today, but by who builds the best error correction, the best software, and the best manufacturing base over the next decade.

If you bet on ecosystems, the United States is the favorite. If you bet on state commitment and speed of execution, China is. The smart money says the race is too close to call — and will remain so until someone builds a quantum computer that does something the world actually needs.