Advancements in Quantum Computing: The Battle Between Tech Giants and Startups
For most of my career, x86 was simply what a computer was. Intel’s architecture ran the desktops, the laptops, the servers, and eventually the Macs. If you’d suggested in 2005 that Apple would one day design its own chips and beat Intel on both performance and battery life, most people in IT would have smiled politely and changed the subject.
Then the M-series arrived and the thing everyone knew to be permanent turned out to be a phase.
That’s the frame I bring to quantum computing. Not because quantum is about to replace your laptop — it isn’t, and it won’t. But because thirty years in this field teaches you that the arrangement that looks eternal is usually just the current one, and the shift is normally underway well before most people notice it.
Quick summary: Quantum computers use qubits instead of bits, which lets them attack certain specific problems in ways ordinary computers can’t. They are not faster computers — they’re different ones, useful for a narrow set of tasks. The race stopped being about qubit counts and became about error rates. And the one part that genuinely affects you is encryption, which is why the migration to quantum-resistant cryptography is already happening.
What makes a qubit different
An ordinary bit is a switch: 0 or 1, one of them, always. Everything your computer has ever done is billions of those switches flipping very fast.
A qubit holds a blend of both states at once — superposition — and only settles into a definite answer when measured. Link qubits together through entanglement and their states become correlated, so operating on one affects the others.
The popular explanation is that this lets a quantum computer “try every answer at once.” That’s not quite right, and the difference matters. You don’t get to read all those answers — measurement collapses everything to one result. The real trick is arranging the calculation so the wrong answers cancel each other out and the right one is overwhelmingly likely to be what you measure.
Which is why quantum computers aren’t general-purpose. That cancellation trick only works for problems with the right mathematical shape. For everything else — email, spreadsheets, video, this webpage — a classical computer is better, and always will be.
Why qubit counts stopped being the headline
For years the coverage was a scoreboard: this lab has 50 qubits, that one has 400, the next has 1,000. It made for clean headlines and told you almost nothing.
Qubits are extraordinarily fragile. They need near-absolute-zero temperatures, heavy shielding, and near-total isolation from vibration and stray electromagnetic noise. Left alone they lose their state in fractions of a second. A machine with a thousand noisy qubits can easily be less useful than one with a hundred stable ones — the errors compound faster than the extra capacity helps.
So the field reorganised around error correction: combining many physical qubits into one reliable “logical” qubit that can survive its components misbehaving. The ratio is brutal — potentially thousands of physical qubits per logical one — which is exactly why a machine advertising a big raw number may still not run a useful algorithm.
When you next see a quantum announcement, that’s the number to look for. Not how many qubits. How many logical qubits, and what the error rate is.
The giants and the specialists
IBM, Google, and Microsoft can fund a decade of research with no product at the end, build custom fabrication, and keep going through the quiet years. IBM has been unusually public about its roadmap, which is useful for the rest of us — it’s a stated timeline that can be checked against reality.
The specialists — IonQ, PsiQuantum, D-Wave and others — bet on one approach and pursue it hard. Trapped ions, photonics, annealing: genuinely different physical strategies for building a qubit, with different failure modes.
That divergence is the healthiest thing about the field. Nobody knows yet which physical approach scales, and a single shared assumption would be far more dangerous than the current spread of bets. One of them being right is enough.
The part that actually touches you
Most quantum applications — molecular simulation, materials discovery, certain optimization problems — are things you’ll benefit from indirectly, if at all, through better drugs or better batteries.
Encryption is the exception, and it’s worth understanding calmly.
Much of today’s public-key encryption rests on multiplication being easy and factoring being hard. A sufficiently large, sufficiently reliable quantum computer running the right algorithm would break that assumption. No such machine exists, and estimates for when one might vary by decades.
The reason people take it seriously anyway is “harvest now, decrypt later” — an adversary recording encrypted traffic today and storing it until a machine exists that can open it. For most of us that’s irrelevant. For state secrets and medical records with a fifty-year sensitivity window, it isn’t.
Which is why the migration is already underway. NIST has standardised post-quantum cryptographic algorithms, and browsers, messaging apps, and operating systems have been quietly adopting them for a while now. This is the rare case where the industry started fixing the problem well before it arrived — and you get it by keeping your software updated, which is the entire action item.
What I’d do: Follow it as an interesting field, install your updates, and treat post-quantum encryption as something already being handled on your behalf.
What I’d skip: Any product marketing itself to consumers as “quantum” anything. Also any headline implying encryption is about to collapse — that story has run every year for a decade and the timeline keeps not arriving.
Never assume this is it
Quantum computing may become genuinely important, or it may stay a specialist tool for a handful of scientific problems. Both outcomes are plausible, and anyone certain which is coming is guessing with confidence.
What I’m confident about is the pattern. x86 looked permanent until it wasn’t. Something will eventually make today’s hardware look like a transitional step, and it may not be quantum at all — it may be something not yet on the list of things being watched.
Ten years ago almost nobody predicted the tools we now use daily. Fifty years ago they wouldn’t have come close. That’s the most reliable lesson technology teaches: never assume this is it, and nothing bigger is coming. Something always is.
Verified resources & documentation
Keep reading
- Why You Can’t Ban AI — and What Encryption’s History Tells Us
- Passkeys in 2026: Why They Are a Smarter Choice Than Traditional Passwords
- On-Device AI vs Cloud AI — Performance, Security, and Apple’s Secure Cloud vs Google
This is general information, not professional advice. Your situation may differ — verify with a professional for high-stakes decisions.