What is quantum computing?
Quantum computing is the technology that has been fifteen years away for about thirty years. That sounds dismissive, but the shape of the field really has changed in the last couple of years, and the reason it keeps coming up is that when it arrives at useful scale it will break some of the assumptions the internet and most of our banking systems are built on. It is worth understanding the basics without getting lost in the physics.
A normal computer stores information in bits. A bit is either a one or a zero. Every Excel spreadsheet, every web page, every photo on your phone, is built from billions of those ones and zeros being flipped very quickly. A quantum computer uses qubits. A qubit can be a one, a zero, or a particular combination of both at the same time. That combination state, called superposition, is where quantum computing gets its strange reputation from, and also where its usefulness comes from.
Superposition on its own is not the whole story. Two qubits can also be linked, a property called entanglement, so that what you do to one instantly affects the other. Combine superposition and entanglement across many qubits and a quantum computer can explore a vast number of possible answers to a problem at once, and let the ones that are useful stand out at the end. It is not magic. It is a very specific kind of cleverness that only works for certain kinds of problems.
The problems it is good at are narrow but consequential. Simulating molecules, which matters for drug design and battery chemistry. Optimising very large logistics puzzles. Searching unsorted spaces faster than classical computers can. And, most famously, breaking certain kinds of encryption, including the kind that secures most of the internet today. The last one is the reason intelligence agencies and banks have quietly been storing encrypted data for years on the assumption that a powerful enough quantum computer will eventually be able to read it.
Where the field actually stands today is the interesting part. The leading machines now have a few hundred to a few thousand qubits. The problem is that qubits are extraordinarily sensitive. Any noise, heat, or stray vibration makes them lose their state, which is why most quantum computers are cooled to near absolute zero and shielded from everything. Error correction, the art of using many physical qubits to build one reliable logical qubit, is the current hard problem. Several teams, including Google, IBM, and UK startup Quantinuum, have recently shown meaningful progress, which is why the conversation has shifted from hypothetical to when.
The UK has a serious stake in this. Oxford and Cambridge are central to the research. The National Quantum Computing Centre sits in Harwell. Quantinuum, formed from the merger of Cambridge Quantum and Honeywell Quantum Solutions, is one of the global leaders in trapped ion quantum computing. The government has committed around two and a half billion pounds over ten years to a national quantum strategy. This is one of the few deep tech areas where Britain is not merely playing along.
A common misconception is that a quantum computer is just a very fast normal computer. It is not. For most of what you do at work and at home, a quantum computer would not be any faster than your laptop, and in many cases it would be much slower. It is a specialist tool for specialist problems. Running Word, streaming Netflix, editing a spreadsheet. None of those need a quantum computer, and none of them ever will.
Another common misconception is that quantum computing will arrive suddenly and break everything overnight. In reality the field is progressing in a bumpy but visible way, and the defensive move for the encryption problem is already in motion. Post quantum cryptography, the branch of mathematics that designs encryption which even a quantum computer cannot crack, has been standardised by the US National Institute of Standards and Technology, and browser makers, banks, and cloud providers are quietly rolling it out. The upgrade will be done before most people notice.
One more thing worth knowing is how the industry actually measures progress. The key metric is not raw qubit count, it is logical qubit count after error correction. A machine with a thousand noisy physical qubits might only run as a handful of logical qubits once the error correction overhead is accounted for. This is why announcements about qubit numbers alone can be misleading. The useful benchmark is how many error corrected operations a machine can perform before it breaks down, and this number, while still small, has been roughly doubling every year in the leading labs.
It is also worth understanding the business model underneath. Most quantum computers today are not sold like laptops. They are run in central facilities by their manufacturers and rented out over the internet by the minute, a model that mirrors early time sharing on mainframes in the 1960s. This is almost certainly how quantum computing will reach ordinary users, through cloud services offered by IBM, Microsoft, Google, AWS, and Quantinuum, rather than through a box you can buy. The UK’s National Quantum Computing Centre is building out a similar access model for British researchers and smaller companies.
The practical takeaway is this. You do not need to learn quantum mechanics. You do need to know that some time in the next five to ten years, this technology will start to matter for drug research, materials science, and cryptography. If you run a business that depends on long lived secrets, medical records, pension data, national security, you should already be asking your providers about their post quantum roadmap. For everyone else, this is a story to watch rather than act on, but it is one of the few genuine long term shifts in computing worth paying attention to.