A different kind of computer
Quantum computing is one of the most hyped and least understood topics in technology. Headlines promise machines that will break all encryption or cure disease overnight. The reality is more subtle and, in its own way, more interesting. A quantum computer is not simply a faster version of the laptop on your desk. It is a fundamentally different kind of machine, built to exploit the strange rules of the very small.
To understand why it matters, you first have to accept that it will not replace ordinary computers for everyday tasks. Instead, it aims to solve a narrow set of problems that are effectively impossible for even the largest traditional supercomputers.
Bits versus qubits
An ordinary computer stores information in bits, each of which is either a zero or a one. Everything your computer does — text, images, video, calculations — is ultimately built from long strings of these two values. It is simple, reliable, and astonishingly powerful when scaled up.
A quantum computer uses quantum bits, or qubits. Thanks to a property called superposition, a qubit can exist in a blend of zero and one at the same time. It is not that we simply do not know its value; the qubit genuinely holds a combination of both possibilities until it is measured. This is where quantum computing gets its unusual power.
The differences are easier to see side by side:
| Aspect | Bits | Qubits |
|---|---|---|
| Basic unit | A bit is 0 or 1 | A qubit is a blend of 0 and 1 |
| States | One value at a time | Many combinations explored at once |
| Scaling | Adding bits adds capacity steadily | Adding qubits grows the space exponentially |
| Error sensitivity | Very robust; errors are rare | Extremely fragile; heat or vibration destroys the state |
| Best use | Everyday computing, from phones to supercomputers | Narrow specialist problems such as simulation, optimisation, and certain mathematics |
Superposition and entanglement
Superposition means that a group of qubits can represent many combinations at once. Where a handful of ordinary bits can hold a single number, the same number of qubits can, in a sense, explore a vast range of numbers simultaneously. Add more qubits and this space grows explosively.
A quantum computer does not try every answer one by one. It uses the physics of interference to make wrong answers cancel out and right answers stand out.
The second key property is entanglement, where qubits become linked so that the state of one instantly relates to another, no matter the distance between them. Entanglement lets a quantum computer coordinate its qubits in ways that have no equivalent in ordinary computing, and it is essential to how quantum algorithms work.
What quantum computers might do
Quantum machines are not general-purpose speed boosters. They excel only at specific kinds of problems where their strange properties give an advantage. The most promising areas share a common feature: an enormous number of possibilities that must be searched or simulated.
- Simulating molecules and materials, which could speed up drug and battery discovery
- Optimising complex systems such as logistics networks or financial portfolios
- Certain kinds of search and mathematics that underlie modern encryption
- Advancing scientific research into chemistry and physics that classical machines model poorly
Notably, simulating the quantum world of atoms and molecules is something classical computers do badly, because nature at that scale is itself quantum. Using a quantum machine to model quantum chemistry is one of the field's most genuinely promising applications.
Encryption is the application that draws the most headlines, and it deserves a careful word. Much of today's online security relies on mathematical problems that ordinary computers cannot solve quickly. A large, reliable quantum computer could, in theory, crack some of these, which is why researchers are already developing new 'post-quantum' encryption designed to resist such machines. Importantly, this threat is years away and is being addressed well in advance, so there is no cause for alarm — but it explains why the field attracts so much attention from governments and security experts.
It is just as important to stress what quantum computers will not do. They will not make your spreadsheets load faster, your videos stream more smoothly, or your everyday apps run better. For the overwhelming majority of computing tasks, an ordinary processor is not merely adequate but superior, because it is cheaper, simpler, and already extremely fast. Quantum machines are specialists for a narrow class of problems, not a general upgrade — a distinction that a great deal of breathless coverage manages to blur. Keeping that distinction clear is the best defence against both hype and disappointment. They could also help design better batteries and materials for renewable energy.
The enormous engineering challenge
Building a useful quantum computer is fiendishly difficult. Qubits are extraordinarily fragile. The slightest heat, vibration, or stray electromagnetic signal can disturb them and destroy the delicate quantum state, a problem called decoherence. To hold qubits still, many machines must be chilled to temperatures colder than deep space.
Errors are the central obstacle. Today's quantum computers make mistakes far too often for serious work, so a huge research effort focuses on error correction — using many physical qubits to build one reliable logical qubit. Progress is real but slow, and practical, large-scale machines remain years away. Most people who use quantum computers today reach them remotely, through the same kind of services described in our guide to cloud computing.
Who is building them
The race has three clear frontrunners, each betting on different technology.
- IBM has the most aggressive public roadmap: its current Nighthawk processor runs 120 qubits, and it plans Starling — its first fault-tolerant machine — for 2029, aiming for 200 qubits capable of 100 million operations.
- Google made headlines with its Willow chip, which solved a thirty-year-old problem: with Willow, adding more qubits reduces errors instead of multiplying them.
- Microsoft and Quantinuum together demonstrated 12 reliable logical qubits — proof that error correction works on real hardware.
A realistic timeline
In about five years, expect steady engineering progress rather than a revolution. IBM aims to demonstrate a practical quantum advantage — a genuinely useful task done better on a quantum machine — by the end of 2026, though it had not done so as of mid-2026. Error correction keeps improving, and post-quantum encryption will quietly roll out across banking.
In about twenty years, the transformative machines may arrive: many researchers expect small fault-tolerant computers within a decade, with the universal fault-tolerant era — thousands of reliable logical qubits — in the 2030s and beyond. That is the timeframe in which quantum chemistry could reshape drug and materials discovery, and in which the long-feared threat to today's encryption would finally become real — which is exactly why cryptographers are replacing vulnerable encryption now, years early.
Nobody can date these milestones precisely, and past predictions have often been optimistic. But the direction of travel is clear.
A realistic view of the future
It is easy to be swept up in the hype or dismiss quantum computing as science fiction. The truth sits between the two. Quantum computers will almost certainly never replace your phone or laptop. But for a narrow band of extremely hard problems, they may one day do things no classical machine ever could.
The most likely future is a partnership: classical computers handling everyday work and quantum machines called upon, like specialist tools, for the rare problems that suit them. Whether that future arrives in ten years or thirty, the science being done today is quietly laying the groundwork for it. Security experts are not waiting, either: systems built on cryptography, from online banking to blockchains, are already being upgraded to resist future quantum attacks.
