JOVANA
Library Glossary Getting Started Three Levels Fields How it works Mission
Join the mission
All guides

What Quantum Is Not

Quantum computing is the most over-hyped topic in tech, and most of what you hear about it is wrong. This guide names the four claims you'll meet most often and corrects each one honestly. Then it lands on what is actually true: a few specific speedups, real but early hardware, and genuine promise in simulating nature.

'Tries all answers at once' — no

The most repeated line about quantum computing is that it "tries every answer at the same time." It's a tempting picture, and it's wrong. A qubit in superposition is not secretly holding many values you can read out. It holds a single quantum state described by amplitudes — numbers that say how much of |0> and how much of |1> are in the mix.

|psi> = a|0> + b|1>,   |a|^2 + |b|^2 = 1
A one-qubit state: a and b are amplitudes, and the squared sizes are the probabilities of reading 0 or 1.

Here is the catch that the hype skips: you only get one measurement. When you read a qubit, the state collapses to a single classical bit, 0 or 1, with a probability set by those amplitudes. You don't get to peek at the whole superposition. So a quantum computer that merely held all answers at once would be useless — you'd just measure one random one.

'Exponentially faster at everything' — no

You'll hear that quantum computers are "exponentially faster." Faster at *what*, though? That qualifier is everything. For the vast majority of tasks — spreadsheets, web browsing, video, most everyday computing — a quantum computer offers no speedup at all, and is in fact slower and far more fragile than the laptop in front of you.

The honest picture is that speedups are specific and problem-shaped. Grover's algorithm searches an unstructured list in about sqrt(N) steps instead of N — a real and useful gain, but a quadratic speedup, not an exponential one. The famous exponential speedups, like Shor's algorithm for factoring large numbers, only show up for problems with special mathematical structure the algorithm can exploit.

'It's just parallel computing' — no

Another tidy mental model: a quantum computer is like a giant parallel machine, running many classical computations side by side. It feels close to right, but it leads you straight back to the "all answers at once" mistake. Classical parallelism gives you many independent results you can each read. A quantum computer gives you one result on measurement — so the win can't come from reading many outputs.

What actually makes quantum different is having no classical equivalent for two of its ingredients: superposition with interference, where amplitudes can be negative or complex and cancel, and entanglement, where qubits share a joint state that can't be described qubit-by-qubit. A pile of parallel classical processors has neither. That's why a quantum computer isn't just "more cores" — it's a different kind of machine, good at a different and much narrower set of jobs.

'Quantum computers are here' — sort of (NISQ)

Real quantum hardware exists — you can run a circuit on one today over the cloud. But "it exists" and "it's ready" are very different claims. We are in the [[nisq|NISQ]] era: Noisy Intermediate-Scale Quantum. The machines have tens to a few hundred physical qubits, and every one of them is noisy — qubits lose their state in microseconds and gates make small errors that pile up over a computation.

There is no large-scale, fault-tolerant quantum computer yet. The fix for noise is quantum error correction — spreading one reliable logical qubit across many physical ones. But schemes like the surface code need error rates below roughly a 1% threshold *and* a heavy overhead — often hundreds to thousands of physical qubits per logical one. We don't have machines anywhere near that scale today.

What is actually true

Clearing away the myths doesn't leave nothing — it leaves something real and worth caring about. Quantum computing is not magic and not universal, but it is a genuine new tool with a few standout uses where the physics is on its side.

  1. A few specific speedups are real. Shor's algorithm would break today's RSA and ECC if a large enough machine existed — which is exactly why post-quantum cryptography is being deployed now, on classical computers, ahead of the threat.
  2. The hardware is real but early. You can run small circuits today, but they're noisy NISQ devices, not the fault-tolerant machines that headline-grabbing algorithms need.
  3. The deepest promise is simulating nature. Using a quantum system to model other quantum systems — molecules, materials, chemical reactions — is where a near-term quantum advantage looks most plausible, since the problem and the machine speak the same physical language.

Hold both halves at once: most quantum hype is wrong, *and* a real, narrow, important technology is being built underneath it. Knowing the difference is the whole point of this rung — it's what lets you read the next quantum headline without being fooled by it.