NISQ (Noisy Intermediate-Scale Quantum)

NISQ, short for Noisy Intermediate-Scale Quantum, is the term coined by physicist John Preskill in 2018 to describe the current generation of quantum machines: devices with dozens to a few hundred qubits that are too noisy to run Quantum error correction at useful scale and too small to threaten deployed cryptography. It names an era, not a specific technology.

#Origin of the term

Preskill introduced NISQ in a 2018 paper surveying the near-term prospects of quantum computing (Preskill 2018). He chose the phrase deliberately. "Noisy" emphasizes that without error correction, imperfect gates and Decoherence limit how deep a Quantum circuit can run before the result is swamped by errors. "Intermediate-scale" points to machines large enough to be hard to simulate classically, yet far smaller than the millions of qubits a fault-tolerant, cryptanalytic computer would need. The paper was candid that NISQ devices might not deliver practical value, framing the era as one of exploration rather than guaranteed payoff.

#What defines a NISQ device

• Tens to hundreds of physical qubits, with no or only rudimentary error correction.
• Gate and measurement error rates high enough to limit circuit depth to at most hundreds to a few thousand operations.
• Reliance on hybrid quantum-classical algorithms, such as variational methods, that try to extract value from short, noisy circuits (Bharti et al. 2022).
• No demonstrated Quantum advantage on any practically useful problem.

The label applies across hardware modalities: Superconducting qubits, Trapped-ion qubits, Neutral-atom qubits, and photonic systems are all NISQ-era today, as are Quantum annealing machines by a similar standard.

#Algorithms for the NISQ era

Because deep circuits are impossible without error correction, NISQ research favors hybrid quantum-classical algorithms that keep the quantum part short. In these methods a classical optimizer repeatedly adjusts the parameters of a shallow Quantum circuit, using measured results to guide the search. The variational quantum eigensolver, aimed at estimating ground-state energies in chemistry and materials, and the quantum approximate optimization algorithm are the most studied examples (Bharti et al. 2022). These approaches are heuristic: they run on present hardware, but whether they offer a scalable advantage over classical methods is unresolved, and noise limits how well they perform as problem size grows.

#Why the distinction matters

The NISQ framing sets honest expectations. It separates two very different questions: whether a machine can do something classically hard on a contrived benchmark, and whether it can do something useful. It also clarifies the cryptographic timeline. Because NISQ devices cannot run Shor's algorithm at cryptographic scale, present hardware poses no near-term threat to RSA or elliptic curve cryptography, and the concern behind post-quantum cryptography rests on future fault-tolerant machines, not current ones.

#Beyond NISQ

Moving past the NISQ era requires crossing into fault tolerance, where encoded logical qubits outperform their physical components. Google's 2024 demonstration of surface-code memory operating below threshold is an early sign of that transition, shown for memory at small scale rather than full computation (Google Quantum AI 2024). As of early 2026 the field is best described as still NISQ, with the first credible steps toward the fault-tolerant era visible but the large, error-corrected machines that would define it not yet built. The distance from here to a cryptographically relevant quantum computer is a central reason the timing of Q-Day cannot be pinned down.