up:: Start Here
What’s a quantum computer?
A machine that’s useless at almost everything and devastating at a very short list of specific problems.
That framing matters more than any description of the physics, because the entire story here comes down to one fact: the mathematics protecting the internet happens to sit on that short list.
The short version:
- An everyday computer works through possibilities one at a time, extremely fast.
- A quantum computer holds many at once and steers them so wrong answers cancel out and the right one remains.
- For almost everything people use computers for, this gains nothing. Quantum computers will never replace laptops.
- For a short list of mathematical problems it’s devastating, and undoing the arithmetic behind encryption is on that list.
- The popular line that it “tries every answer at once” is wrong, and cryptographers reject it.
- Today’s machines have hundreds to low thousands of error-prone parts, where breaking real encryption needs thousands of stable ones.
How does it compare to an everyday computer?
| Ordinary computer | Quantum computer | |
|---|---|---|
| How it works | Tests possibilities one at a time | Holds many at once, cancels the wrong ones |
| Good at | Essentially everything you do daily | A short list of specific math problems |
| Bad at | The problems on that short list | Essentially everything you do daily |
| Will you own one? | You already do | No, and there’d be no reason to |
| Threat to encryption | None | Breaks the half that proves identity and sets up privacy |
How the difference actually works
An everyday computer works through possibilities one at a time. It’s extraordinarily fast at that, and it’s still one after another, checking, discarding, moving on.
A quantum computer works on a different principle. It holds many possibilities at once and manipulates them so that the wrong answers cancel each other out while the right answer reinforces itself. What comes out at the end is the answer that survived.
That cancellation is the whole trick, and it only works when the problem has a particular mathematical shape. Most problems don’t.
What’s the description everyone gets wrong?
That a quantum computer “tries every answer simultaneously.”
You’ll see it in almost every news article, and it’s wrong. If it worked that way, a quantum computer would be fast at everything, and the entire field would look completely different.
What actually happens is the cancellation described above. The machine doesn’t check every possibility and pick the winner. It arranges things so the wrong possibilities destroy each other, leaving the right one standing. That distinction is why quantum computers are useless for most tasks, and it’s the first thing a cryptographer will correct in anything written about this.
So what is it actually good at?
A short and specific list, which includes simulating molecules and chemistry, certain optimization problems, and undoing a particular kind of arithmetic.
That last one is the problem. The arithmetic in question is the easy-forwards, impossible-backwards math that keeps encryption keys secret, described at What is encryption? A quantum computer can run that in reverse efficiently, and everything else about this subject follows from that single fact.
Worth being clear about what stays safe: your laptop, your phone’s speed, your spreadsheets, your video calls, and essentially every everyday computing task are unaffected. Nobody is replacing anything you own.
How big are they now, and how big do they need to be?
Today, quantum computers have on the order of hundreds to low thousands of qubits, and those qubits are error-prone. They drift, they pick up noise, and they lose their state easily.
To break the encryption in question, you need thousands of stable qubits, each one built out of many error-prone ones working together to correct each other’s mistakes. That’s the part nobody has solved at scale.
Today’s machines are orders of magnitude short. That’s the honest position, and anyone telling you otherwise is overstating it.
But the requirement has been moving. In 2019, the published estimate for breaking a standard encryption key was 20 million error-prone qubits running for 8 hours. By 2025 that estimate had fallen to under 1 million. The hardware didn’t shrink; the methods improved.
Source: C. Gidney and M. Ekerå, “How to factor 2048 bit RSA integers in 8 hours using 20 million noisy qubits,” Quantum 5, 433, 2021, arxiv.org; C. Gidney, “How to factor 2048 bit RSA integers with less than a million noisy qubits,” 2025, arxiv.org.
A roughly 20-fold drop in the requirement over 6 years is the reason people who work on this stopped treating the arrival date as comfortably distant.
Why is a qubit count a bad way to judge a machine?
Because it’s the number most likely to appear in a press release and the least likely to indicate capability.
A machine with many noisy qubits can be less capable than one with fewer clean ones. What matters more is the error rate, how long the qubits hold their state, and whether they’re error-corrected. A headline announcing a record qubit count often says very little about whether anything got closer.
If you’re reading a story about a new machine, the useful question is what problem it actually solved, and whether anybody outside the company reproduced it.
Questions people ask
Will I ever own one? No, and there’d be no reason to. They’re useless for the things personal computers do.
Do they exist at all? Yes. Working quantum computers exist and are being improved steadily. What doesn’t exist is one large and stable enough to break real encryption.
Has one broken encryption yet? No. There’s no public evidence of a quantum computer breaking a real encryption key. Demonstrations that factor very small numbers recur every few years and get reported as breaks each time.
Is quantum computing the same as quantum encryption? No, and this confusion is everywhere. Quantum key distribution is a separate technology using physics to distribute keys, and it’s rarely what any of this is about. The replacement encryption being deployed today runs on completely everyday computers.
Who’s building them? Several large technology companies, a number of startups, and multiple national programs. Their public claims deserve reading carefully, since the companies raise capital on the machine seeming close.
Where to go next
- When is this going to happen? covers what experts actually predict, with the numbers.
- What is encryption? covers what the machine would break.
- Is this overhyped? states the strongest argument that none of this will arrive.
Go deeper into the technical detail
The technical version is Quantum Computing MOC and Shor’s Algorithm.
These open the Post-Quantum Field Guide, a separate site written for security professionals.
Last verified 2026-07-30 · Maintained by Addie LaMarr, LaMarr Labs.