Science News & Breakthroughs Featured May 23, 2026

Understanding Quantum Computing: A Beginner-Friendly Guide

A careful guide to qubits, interference and entanglement, with a clear line between demonstrated results, such as error-correction milestones, and speculative claims about quantum speedups.

SC
ScienceTrace Science Desk
•
 7 min read
•
 1,322 words

Quantum computing is often described as a machine that tries every answer at once. That picture is popular and misleading, and it hides the interesting part: a quantum computer does not simply run many classical computations in parallel. It manipulates amplitudes, and useful speedups appear only when an algorithm arranges for the wrong answers to cancel out. This guide explains the core ideas with the mathematics kept light, and distinguishes what has been demonstrated from what remains aspiration.

Qubits and what superposition really means

A classical bit is 0 or 1. A qubit's state is a vector in a two-dimensional complex space:

|ψ⟩ = α|0⟩ + β|1⟩, with |α|² + |β|² = 1

The numbers α and β are complex amplitudes. When the qubit is measured, the result is 0 with probability |α|² and 1 with probability |β|², and the state then collapses to the outcome observed. So a qubit "holds both values" only in the sense that its description includes both amplitudes; a measurement still returns a single bit. With n qubits, the state needs 2n amplitudes, which is why simulating a large quantum system on classical hardware becomes expensive. But reading out all those amplitudes is not possible, so the exponential size of the state does not translate directly into exponential computing power.

Interference: where the speedup comes from

Amplitudes can be positive, negative or complex, so they can add up or cancel. Quantum algorithms are designed so that amplitudes of incorrect answers interfere destructively and those of correct answers constructively, making the right answer likely when the final measurement is made. Designing such an algorithm is hard, which is the reason only a modest number of problem families have known quantum speedups.

Entanglement

Entanglement describes states of several qubits that cannot be written as separate states of each qubit. Measuring one qubit of an entangled pair gives outcomes correlated with the other, in ways no local classical model reproduces, as tests of Bell inequalities have shown. It is sometimes described as the qubits "influencing each other instantly", but that wording invites a wrong conclusion: entanglement cannot be used to send a message faster than light, because each party's individual outcomes look random until the results are compared through an ordinary channel. For computing, entanglement is a resource that lets the amplitudes of a multi-qubit register carry correlations a classical register of the same size cannot represent efficiently.

What quantum computers are known to be good at

  • Simulating quantum systems. Richard Feynman suggested in 1982 that quantum systems should be simulated by quantum machines. Estimating the energies of molecules and materials is the most widely discussed application, and resource studies, such as Reiher et al. (2017) on the nitrogenase cofactor, estimate what a fault-tolerant machine would need for chemically relevant problems.
  • Factoring and discrete logarithms. Shor's algorithm (1994) factors integers in polynomial time, which would break RSA and elliptic-curve cryptography if run at scale. Gidney and Ekerå (2021) estimated that factoring a 2,048-bit RSA integer would require on the order of twenty million noisy physical qubits and about eight hours under their assumptions, far beyond today's devices.
  • Unstructured search. Grover's algorithm (1996) finds a marked item among N in about √N steps, a quadratic, not exponential, speedup.

For many other claimed applications, including optimisation, finance and machine learning, the evidence of practical advantage is much weaker. Loading classical data into a quantum state can cost as much as the computation saved, and for some recommendation-system problems Tang (2019) showed that efficient classical algorithms using similar sampling assumptions exist, which removed the claimed exponential advantage there. Statements such as "quantum AI will make AI smarter" are, at present, speculation.

Hardware and noise

Several physical platforms are being pursued: superconducting circuits, trapped ions, neutral atoms, photons and spin qubits. Superconducting devices operate in dilution refrigerators at around ten millikelvin to suppress thermal noise; trapped-ion systems use vacuum chambers and lasers, not extreme cooling of the whole machine. All platforms face decoherence, the loss of quantum information through interaction with the environment, and imperfect gates. Current devices are what Preskill (2018) called NISQ, noisy intermediate-scale quantum machines, with tens to a few thousand qubits and error rates too high to run deep circuits.

Quantum error correction addresses this by encoding one logical qubit across many physical ones. Shor proposed the first code in 1995, and the surface code is a leading candidate because it tolerates relatively high physical error rates. In 2023 a Google team reported a surface-code logical qubit whose error rate improved when the code distance was increased from 3 to 5, an important experimental step showing that scaling up can reduce errors, though it is far from the many reliable logical qubits that large algorithms require.

The "quantum advantage" debate

In 2019, Arute and colleagues reported that a 53-qubit superconducting processor sampled the output of a random circuit far faster than they estimated a classical supercomputer could. The claim was contested: IBM researchers argued that better classical methods could reduce the classical time substantially, and subsequent classical simulation work narrowed the gap. The episode is useful for method. Random circuit sampling is a task chosen to be hard for classical machines and easy for the quantum device, with little practical use, and any advantage depends on the best available classical algorithm, a moving target.

Cryptography is already responding

Because encrypted data can be recorded now and decrypted later, standards bodies have not waited for large quantum machines. In August 2024 the US National Institute of Standards and Technology published its first post-quantum cryptography standards (FIPS 203, 204 and 205), covering key encapsulation and digital signatures based on problems not known to be vulnerable to Shor's algorithm. Quantum technology also offers quantum key distribution, though it addresses a narrower threat model and has its own deployment limits.

Will quantum computers replace classical ones?

No. Everyday tasks such as browsing, word processing and gaming gain nothing from quantum hardware, and a quantum computer is expected to act as an accelerator for particular subroutines, controlled by classical computers that also perform error-correction decoding in real time. The analogy sometimes made to early internet days is only partly apt: the barrier is not market adoption but unresolved physics and engineering.

Limitations and open questions

  • Scale of error correction. Estimates for useful algorithms need thousands of logical qubits, each requiring many physical qubits. Whether overheads can be reduced by better codes is an active research area.
  • Useful advantage. No broadly accepted demonstration yet shows a quantum computer outperforming the best classical method on a commercially or scientifically valuable task.
  • Algorithms. The list of problems with proven quantum speedups remains short, and finding new ones is as much a bottleneck as building hardware.
  • Verification. Checking that a quantum computer produced the right answer for a problem classical machines cannot solve is itself a research problem.

What researchers should watch

  • Logical error rates that keep falling as code size grows, over multiple rounds and with real-time decoding.
  • Resource estimates for chemistry and materials problems, and whether classical methods (tensor networks, density functional theory improvements) keep pace.
  • Claims of advantage that specify the best classical baseline and the time and energy spent on it.
  • Migration to post-quantum cryptography in the systems that protect long-lived data.

References and further reading

  1. Nielsen M. A., Chuang I. L. Quantum Computation and Quantum Information. Cambridge University Press.
  2. Feynman R. P. Simulating physics with computers. International Journal of Theoretical Physics, 1982.
  3. Shor P. W. Polynomial-Time Algorithms for Prime Factorization and Discrete Logarithms on a Quantum Computer. SIAM Journal on Computing, 1997 (first presented 1994).
  4. Grover L. K. A fast quantum mechanical algorithm for database search. Proceedings of the 28th ACM Symposium on Theory of Computing, 1996.
  5. Preskill J. Quantum Computing in the NISQ era and beyond. Quantum, 2018.
  6. Arute F. et al. Quantum supremacy using a programmable superconducting processor. Nature, 2019.
  7. Google Quantum AI. Suppressing quantum errors by scaling a surface code logical qubit. Nature, 2023.
  8. Gidney C., Ekerå M. How to factor 2048 bit RSA integers in 8 hours using 20 million noisy qubits. Quantum, 2021.
  9. Tang E. A quantum-inspired classical algorithm for recommendation systems. STOC, 2019.
  10. National Institute of Standards and Technology. FIPS 203, 204 and 205 (post-quantum cryptography standards), 2024.
#quantum computing #AI #science #technology #qubits
All Scientific Breakthroughs