Quantum computing is an approach to computation that processes information using the rules of quantum mechanics rather than those of everyday physics. Its basic unit is the qubit, which can hold a combination of 0 and 1 at the same time and can be linked, or entangled, with other qubits. For a narrow set of problems, carefully designed quantum algorithms can reach an answer in far fewer steps than any known method on a classical computer.
Quantum computers are not faster general-purpose machines and are not expected to replace ordinary computers for most tasks. Their importance for finance comes from two directions: possible speed-ups in simulation and optimisation, and the threat that a large machine would pose to the public-key cryptography protecting banking systems and blockchains. As of November 2026, existing machines were still experimental and too small and error-prone to break real-world encryption.
Quantum computing in plain terms
Think of searching a huge hedge maze for the exit. An ordinary computer is like one walker trying paths one after another, very quickly. A quantum computer is more like sending a ripple of water through the whole maze at once and arranging things so that the ripples along wrong paths cancel each other out, while the ripple along the right path grows stronger.
The trick only works for mazes with a particular kind of structure, and the water is easily disturbed by any outside noise. That is why quantum computers are extraordinary at a few problems, useless for most, and very hard to build.
How it works
Three effects give quantum computers their power:
- Superposition. A qubit can be in a blend of 0 and 1, and a group of qubits can represent many combinations at once.
- Entanglement. Qubits can be correlated so that measuring one reveals information about others, letting operations act on the group as a whole.
- Interference. Algorithms are designed so that wrong answers cancel and right answers reinforce, making the correct result likely when the qubits are finally measured.
Measurement collapses each qubit to a plain 0 or 1, so a quantum program must be built to leave the useful answer standing at the end. Qubits are physically made from superconducting circuits kept near absolute zero, trapped ions held by electric fields, particles of light or neutral atoms. All are vulnerable to noise, so a central engineering goal is quantum error correction, which spreads one reliable "logical" qubit across many physical ones.
History
The physicist Richard Feynman suggested in the early 1980s that quantum systems could be simulated efficiently only by machines that were themselves quantum. In 1994 Peter Shor published an algorithm that could factor large numbers efficiently on a quantum computer, showing that RSA and similar schemes would be breakable on a large enough machine. In 1996 Lov Grover published a search algorithm offering a more modest speed-up. In 2019 Google reported that its Sycamore processor had completed a narrow sampling task far faster than a classical supercomputer could, a claim of "quantum supremacy" that some researchers disputed. In December 2024 Google described its Willow chip, reporting that errors fell as more qubits were added to an error-correcting code, an important technical milestone. IBM, Quantinuum, IonQ and others have published their own roadmaps.
Potential uses in finance
Banks and asset managers have run research programmes on quantum methods for several tasks:
- Simulation. Quantum algorithms could in theory speed up the Monte Carlo simulations used for derivatives pricing and risk measurement.
- Optimisation. Problems such as optimising portfolio allocation and routing payments through networks are candidates, though clear advantages over classical methods have not been demonstrated at scale.
- Machine learning. Researchers are testing quantum approaches to pattern recognition, including fraud detection, with uncertain results.
Most of this work remains experimental, and claims of near-term commercial advantage should be treated with caution.
The threat to cryptography and crypto
The best-understood impact is negative. A large, error-corrected quantum computer running Shor's algorithm could derive private keys from public keys, undermining the digital signatures that authorise bank messages, card transactions and every bitcoin transfer. The hypothetical day this becomes possible is known as Q-Day. Hash functions and proof-of-work mining are far less exposed, since Grover's algorithm gives only a square-root gain. In August 2024 the US National Institute of Standards and Technology published its first post-quantum cryptography standards so that organisations could begin replacing vulnerable algorithms.
Limitations
Current machines have limited numbers of qubits, short coherence times and high error rates. Useful error correction needs many physical qubits per logical qubit, and estimates of the size needed to break modern encryption run into very large numbers. Progress is real but uneven, and announcements by companies are sometimes framed in ways that overstate practical capability. Timelines for useful or cryptographically relevant machines remain uncertain.