Most security problems announce themselves. A suspicious login, a port left open, a file where it shouldn’t be. You get a signal, and usually a chance to respond.
Quantum computing doesn’t work that way. There’s no alert, no patch, and no log entry. One day, the math that quietly protects a lot of everyday technology — secure websites, signed software, encrypted messages — will simply stop being reliable. It’s worth understanding what that means before it happens, because the fix can’t be rushed at the last minute.
What makes quantum computers different#
You don’t need to understand the physics. The idea that matters is small: these machines don’t work through problems one step at a time the way normal computers do.
Think of a maze. A normal computer walks one path at a time — down a corridor, back up, try the next — until it finds the way out. A quantum computer explores every path at once and settles on the answer.
For most tasks, that difference is only interesting. But for the specific math that protects most of our online security — the kind behind the padlock in your browser, digital signatures, and secure messaging — it’s a serious problem. Problems that would take a normal computer billions of years to solve become quick work for a large enough quantum machine.
When might this happen?#
The point at which a quantum computer becomes strong enough to break today’s encryption is often called Q-Day. Estimates vary widely — some say the 2030s, others the 2040s.
The exact date matters less than it seems. Think about how long the data you protect needs to stay private. A credit card number might only matter for a few years. A patient’s medical record, a legal contract, or a government document may need to stay confidential for decades. If you hold that kind of information, the timeline is already relevant to you — whichever forecast turns out to be right.
A problem that’s already started#
There’s a second part to the risk that’s easy to miss: encrypted data can be saved today and unlocked later. An attacker doesn’t need a working quantum computer to begin. They can collect encrypted information now and hold onto it until the tools catch up.
This has a name — harvest now, decrypt later — and it’s already happening. The idea is straightforward: intercept encrypted traffic or copy encrypted files today, store them, and decrypt them once quantum computers are available. Re-encrypting your data now doesn’t undo it, because the old copy is already out there.
It changes how we should think about encryption. Most of us treat it as “turn it on and forget it” — data encrypted, job done. But if data needs to stay private for a long time and was protected with today’s methods, it becomes a problem waiting for the right tool to arrive.
What gets collected is predictable: anything that keeps its value over time.
- Medical records and insurance details — relevant for a person’s whole life.
- Legal documents and contracts — often significant for decades.
- Financial records — payment details, account information, transaction history.
- Everyday logins and messages — credentials and private conversations.
One area worth particular attention is digital certificates — the system that confirms a website is really what it claims to be, and that software updates come from the people they say they do. If an attacker can forge these, they could impersonate legitimate services or push out fake updates that look genuine. That’s a much bigger deal than any single stolen password.
What actually breaks#
It’s easy to see this as a privacy problem: encrypted data becomes readable, so secrets get exposed. That’s part of it, but not the most important part.
The bigger issue is trust. Digital signatures prove who sent a message. Certificates prove you’re talking to the real website. Code signing proves software hasn’t been tampered with. These are the mechanisms that let people do business online without meeting face to face.
When they stop being reliable, you don’t just get a data leak. You get:
- Documents that can’t be verified. A contract, a court filing, or a notarised record could have been altered — and there’d be no cryptographic way to prove it either way.
- Fake websites that look genuine. The certificate check passes, so there’s no warning.
- Software updates that can’t be trusted. The signature no longer proves the update is what the vendor actually shipped.
To be clear: none of this is happening at scale today. The point is that the raw material for these problems — encrypted traffic, certificates, signed files — is being collected now, so the pieces are in place when the capability arrives.
Where to start#
The practical response isn’t panic. It’s a checklist.
Start by finding where you rely on the kind of encryption that’s at risk — the public-key cryptography behind certificates, signatures, and secure connections. Then identify which of those things would still matter if they were compromised five or ten years from now. Those are the ones to address first.
The rest of this series walks through that migration: which replacement algorithms NIST has standardised, how to test them in your own systems, and where the practical pitfalls are. Part 2 covers the algorithm landscape — what’s been chosen, what’s been broken, and which ones to bet on.
The short version: the risk is real but manageable. The work is to find the things that need protecting long-term, and start moving them to quantum-resistant encryption before the old methods stop working.
