What Quantum Computers Are Actually Good At
Quantum computers are special-purpose accelerators, not faster PCs — great for a short list of structured problems, useless for most of the rest.
After Grover and Shor it’s tempting to picture a quantum computer as just a “faster computer” you’d want for everything. The truth is sharper and far more useful: there’s a short list of things they’re genuinely great at, a medium list where they help a bit, and a huge list where they offer nothing at all. Telling these apart is the difference between hype and judgment — and it’s the most practically valuable thing this whole track can give you. Hold the question: what actually makes a problem a good fit for a quantum computer?
Three tiers of quantum advantage
Sort the known algorithms into three buckets. Exponential wins: problems with deep hidden structure — factoring/period-finding (Shor) and simulating quantum systems (chemistry, materials). These are the game-changers. Quadratic wins: amplitude amplification (Grover) and its many uses inside search and optimization — a real but modest √N boost, broadly applicable. No known win: the vast majority of everyday computing. Most problems simply don’t move to a quantum machine.
The common thread: exploitable structure
Every speedup traces back to one thing — structure that interference can latch onto. A hidden period, a quantum system’s own dynamics: these let amplitudes for wrong answers cancel systematically and the right answer reinforce, sometimes exponentially. Strip the structure away (pure unstructured search) and the best you can do is Grover’s quadratic nudge. Generic, structureless problems give interference nothing to work with — which is exactly why “faster at everything” is false.
So what are quantum computers FOR?
An honest map. Chemistry & materials simulation — likely the biggest, perhaps earliest, real payoff. Cryptanalysis (Shor) — a security threat driving post-quantum crypto, not a consumer feature. Some optimization and search subroutines — quadratic help, useful but not revolutionary. And not: spreadsheets, web serving, ordinary apps, most number-crunching. A quantum computer is a special-purpose accelerator — like a GPU is for graphics — that sits beside classical computers for specific jobs, not a replacement for your laptop.
A deep-sea submarine, not a faster car. A submarine is extraordinary in one environment — the deep ocean — and useless for the daily commute; buying one to beat traffic would be absurd. A quantum computer is the same: superb for a few structured problems (simulating molecules, factoring) and pointless for ordinary computing. Using one to run your email would be like driving a submarine to the office. The right question is never “is it more powerful?” but “is this the ocean or the highway?”
Classify five problems by asking “is there structure interference can exploit?”: 1. Factor a 2048-bit RSA key → hidden periodic structure → Shor → exponential win (given a big enough machine). Great fit. 2. Simulate a catalyst molecule → a quantum system, native to qubits → exponential win. Great fit. 3. Find one record in an unsorted database of N → no structure → Grover → ~√N, a modest win. OK fit. 4. Sort a list / run a spreadsheet / serve a web page → no exploitable quantum structure → no speedup. Bad fit — use a classical computer. 5. The pattern: deep structure → maybe exponential; search-like → quadratic; none → classical wins. That one question is your filter for every quantum claim you’ll ever hear.
This is the reading. The interactive version — active-recall quiz, a hands-on experiment you run in your own AI, and an earned mastery check — is free in the app.
Start this lesson free →