MODULE 6 · LESSON 3
Free — no login requiredSign in to track progress, save quiz attempts and enrol in the full course.
Sign in to track progress / enrolThe Honest Summary
A closing lesson that collects the course into something you can state, and defend, in a meeting.
What is settled
These are not seriously disputed by anyone informed.
- The physics works. Superposition, interference and entanglement are established, and quantum computers do what the theory says on the scales tested.
- Shor's algorithm breaks RSA and elliptic curve cryptography given a sufficiently large error corrected machine. The mathematics is proven.
- Quantum error correction works in principle, and following the below threshold demonstrations of the last two years, in practice at small scale.
- Simulating quantum systems is the strongest application. It follows from the structure of the problem rather than from optimism.
- Post-quantum cryptography is necessary and available, and the migration should already be underway.
What is contested
Reasonable, informed people disagree here, and anyone expressing certainty is overreaching.
- When a cryptographically relevant machine arrives. Credible estimates span roughly a decade to several decades, and the spread reflects genuine uncertainty about engineering scaling rather than ignorance.
- Which hardware approach wins, or whether several coexist.
- Whether NISQ machines will deliver commercial value before error correction matures, or whether the hybrid era is a detour.
- How large the optimisation opportunity really is. Researchers are broadly sceptical, some practitioners report useful results, and the evidence remains thin.
What is genuinely unknown
- Whether new algorithms will substantially widen the application list. The list has grown slowly for three decades. It might broaden, and there is no way to predict it.
- Whether an unexpected obstacle to scaling exists. Nothing known forbids large fault tolerant machines. That is not the same as knowing none exists.
- Whether post-quantum algorithms will themselves hold. Lattice based cryptography is well studied and not proven secure, which is why the standards include a hash based alternative resting on different assumptions.
Why both extremes are wrong
The dismissive position, that this is hype and nothing will come of it, ignores that the physics is demonstrated, that error correction has now been shown to work below threshold, that serious money and serious scientists are engaged, and that the cryptographic threat is real enough that every major standards body has acted on it.
The breathless position, that quantum computers will shortly revolutionise everything, ignores that the useful algorithm list is short, that the gap between hundreds of noisy qubits and millions of error corrected ones is enormous, that a quadratic speedup does not defeat exponential growth, and that most advertised applications have no cancellation mechanism behind them.
The defensible position sits between, and it is specific rather than merely moderate:
Quantum computing is a real technology with a narrow and important set of applications, arriving on a timescale of decades rather than years, whose only urgent consequence for most organisations is a cryptography migration that should already be underway.
That sentence is worth being able to say. It is accurate, it is not evasive, and it will save your organisation money in both directions.
What to watch
If you follow anything, follow these rather than qubit counts.
Logical qubit counts and error rates, not physical qubit counts. Progress in fidelity often matters more than progress in scale, because overhead depends steeply on error rates.
Error correction overhead. If approaches such as qLDPC codes deliver an order of magnitude reduction in physical qubits per logical qubit at scale, timelines compress substantially. This is the most likely source of a genuine surprise.
Results in your own domain, if you are tier two. General progress matters less than progress on problems shaped like yours.
Classical algorithm improvements. Dequantisation results are as informative as quantum ones, and they receive far less coverage.
Post-quantum migration guidance from your national body, which affects you regardless of tier.
A closing observation that generalises beyond quantum computing.
The field has an unusual property: the science is genuinely excellent and the commercial claims are frequently poor, and both are true at once. That combination confuses people, because we normally use commercial noise as a signal that the underlying thing is empty, and here that inference fails.
Shor's algorithm was published in 1994. More than three decades later there is still no machine that can run it usefully. In most technologies that would indicate a dead end. Here it indicates a genuinely hard engineering problem being worked on steadily, with real milestones passed along the way, of which below threshold error correction is the most recent and most significant.
Two practical consequences follow for anyone in an organisation.
Do not extrapolate from press releases. The gap between demonstration and deployment in this field is measured in years to decades, and a working laboratory result does not imply a product. Module 4's checklist exists for this reason.
Do not conclude that slow means fake. The steady, unglamorous progress in error rates and logical qubits is the real story, and it is the part least covered because it does not produce good headlines. A team announcing a modest fidelity improvement may have advanced the field more than one announcing a larger chip.
The organisations that will handle this well are not the ones that move first or the ones that dismiss it. They are the ones that do the cryptography migration on schedule, keep one person genuinely informed, and remain able to recognise a real result when one arrives. That is an inexpensive posture, and it is available to almost anyone.
You now have what you need to hold it.
Which statement best represents a defensible position on quantum computing?
Settled
Click to flipThe physics, Shor's mathematics, error correction in principle and now below threshold, chemistry as the strongest application, and the need for post-quantum cryptography.
Click to flip backSeparate the settled from the contested from the unknown. Settled: the physics, Shor's mathematics, error correction working below threshold, chemistry as the strongest application, and the need to migrate cryptography. Contested: timelines, winning hardware, whether NISQ pays. Unknown: whether the algorithm list broadens. Both extremes are wrong, and the defensible position is specific: a real technology with narrow important applications on a timescale of decades, whose one urgent consequence is a cryptography migration already overdue. Watch logical qubits, error rates and overhead, not physical qubit counts.