Self-Assessment Quiz: The Quantum Future

Twenty questions on what comes after NISQ: fault tolerance, quantum networks, and how to think about the field's trajectory. Aim for 16+.

Question 1

The transition from NISQ to fault tolerance is gated primarily by:

A) Logical qubit count and logical gate fidelity B) Physical qubit count alone C) Classical software D) Funding

Question 2

The first commercially valuable fault-tolerant application is most likely to be:

A) Quantum simulation of correlated materials and chemistry B) Combinatorial optimization C) Machine learning on classical data D) Cryptanalysis

Question 3

Cryptanalysis is unlikely to be first because:

A) It requires far more logical qubits and $T$ gates than simulation B) The algorithm does not work C) It is illegal D) Nobody wants it

Question 4

A quantum network's core primitive is:

A) Distributed entanglement B) Classical messaging C) Photon counting D) Encryption

Question 5

Quantum repeaters are needed because:

A) No-cloning forbids amplifying quantum signals B) Fibre is too expensive C) Photons travel too slowly D) Detectors are inefficient

Question 6

Distributed quantum computing would allow:

A) Multiple modules to act as one larger computer via entanglement links B) Cloud access C) Classical parallelism D) Faster classical networking

Question 7

The main obstacle to distributed quantum computing is:

A) Low rate and fidelity of inter-module entanglement B) Software C) Cost D) Regulation

Question 8

Quantum sensing differs from computing in that it:

A) Is already commercially deployed in several applications B) Requires more qubits C) Is purely theoretical D) Needs error correction

Question 9

A blind quantum computing protocol would let a user:

A) Run a computation on a remote server without revealing the computation B) Compute without qubits C) Break encryption D) Avoid measurement

Question 10

Quantum advantage in machine learning is most plausible for:

A) Learning on quantum data B) Image classification C) Language models D) Tabular data

Question 11

The most likely near-term societal impact of quantum computing is:

A) The forced global migration to post-quantum cryptography B) Drug discovery C) Climate modelling D) Financial optimization

Question 12

"Post-NISQ" refers to:

A) The era of early fault-tolerant devices with modest logical qubit counts B) Abandoning quantum computing C) Classical simulation D) Quantum annealing

Question 13

Early fault-tolerant algorithms are being designed to:

A) Use few logical qubits and low $T$-counts, trading runtime for resources B) Maximize qubit usage C) Avoid error correction D) Run on NISQ hardware

Question 14

The most under-appreciated component of a fault-tolerant machine is:

A) The real-time classical decoder B) The refrigerator C) The qubits D) The user interface

Question 15

Progress is best tracked by:

A) $\Lambda$, verified logical qubit count, and logical gate error B) Physical qubit count C) Funding raised D) Press coverage

Question 16

True or false: Quantum computers will eventually replace classical computers for general workloads.

Question 17

True or false: A quantum internet would primarily be used for faster data transfer.

Question 18

True or false: Progress in codes and algorithms has moved resource estimates more than hardware has.

Question 19

Short answer. Describe what a realistic "useful quantum computer" looks like in practice.

Question 20

Short answer. What is the most important thing to carry forward from this book?

Answer Key

Q Ans Note
1 A Physical qubits without error correction do not compose into useful computation.
2 A Native quantum input, exponential classical baseline, and the lowest resource requirements of the candidate applications.
3 A RSA-2048 needs ~6,000 logical qubits and $10^{10}$ $T$ gates; useful chemistry needs far less.
4 A Entanglement distribution, from which teleportation, QKD, and distributed computation follow.
5 A Amplification of an unknown quantum state is forbidden, so loss cannot be compensated classically (Ch. 9).
6 A Modules linked by entanglement act as one machine — the route past single-refrigerator limits.
7 A Inter-module links are orders of magnitude slower and noisier than intra-module gates.
8 A Quantum sensors — magnetometers, gravimeters, atomic clocks — are commercial products today, unlike quantum computers.
9 A Blind and verifiable delegated computation, which matters commercially once cloud quantum computing has value worth protecting.
10 A No data-loading bottleneck and proven sample-complexity separations (Ch. 21).
11 A It is already happening, is certain, and affects every organization — unlike every other candidate impact.
12 A Early fault tolerance: tens to hundreds of logical qubits, not thousands.
13 A Trading longer runtimes for fewer logical qubits, since qubits are the scarce resource first.
14 A Terabit-scale real-time decoding is a major unsolved classical engineering problem (Ch. 29).
15 A The three numbers that carry information.
16 False They are accelerators for specific structured problems. Classical computers remain superior for essentially everything else, and always will.
17 False A quantum internet distributes entanglement, enabling QKD, distributed computing, blind computation, and networked sensing. It does not move classical data faster — classical fibre already does that better.
18 True The RSA-2048 estimate fell ~50× from algorithmic and error-correction advances with no hardware improvement (Ch. 15), and qLDPC codes offer a further ~24× (Ch. 23).
19 — Not a desktop machine and not a replacement for anything. A room-sized cryogenic system, or a network of them, accessed over the cloud, containing perhaps hundreds to thousands of logical qubits built from millions of physical ones, attached to a large classical decoding and control system that consumes more power than the quantum processor. It runs specific simulation and chemistry workloads that classical computers cannot, as an accelerator within a classical pipeline — most of the work before and after remains classical.
20 — The habit of verification. Every recurring failure in this field — strawman baselines, histograms presented as proof of entanglement, missing error bars, mitigated-only reporting, qubit counts without fidelities, query complexity mistaken for gate complexity — is a verification failure. The physics is sound and the engineering is genuinely hard; what separates useful work from noise is whether claims are checked against the right baseline, with honest uncertainty, and with an explicit statement of what was and was not shown. That discipline outlasts any particular algorithm or platform.