Drawing set · Quantum technology
Recent breakthroughs in quantum technology, worked through carefully
bddyreelsmediabreez reads the published record on quantum computing, quantum-safe cryptography and quantum sensing, and tries to separate what has been demonstrated from what has merely been announced. Where the evidence is thin, we say so rather than round up.
Contents · Five sheets
Where the field has actually moved
Each sheet takes one strand of the subject and works through the trade-offs. The order runs from the deepest engineering problem to the claims that get the most attention, because the second only make sense once you have the first.
- Sheet 01
Quantum error correction crosses the threshold
Why the surface-code results from Google Quantum AI, the neutral-atom logical qubits from the Harvard and QuEra collaboration, the Quantinuum and Microsoft trapped-ion work and IBM's low-density parity-check codes matter more than raw qubit counts, and where "logical qubit" is doing more work than it should.
Read sheet 01 → - Sheet 02
Superconducting, trapped-ion, neutral-atom and photonic hardware
A comparison of the platforms pursued by IBM, Google, Quantinuum, IonQ, QuEra, Pasqal, Atom Computing, PsiQuantum and Xanadu on gate speed, connectivity, coherence and the practical problem of wiring and cooling a machine that has to grow.
Read sheet 02 → - Sheet 03
Quantum-safe cryptography after the NIST standards
What Shor's algorithm breaks and what it leaves alone, the ML-KEM, ML-DSA and SLH-DSA standards NIST published in 2024, the UK National Cyber Security Centre's migration guidance, and why quantum key distribution from the Micius satellite is a separate answer to a narrower question.
Read sheet 03 → - Sheet 04
Quantum sensing, optical clocks and gravimetry
Optical lattice clocks and the road to redefining the SI second, nitrogen-vacancy diamond magnetometers, the University of Birmingham gravity-gradient detection of a buried tunnel, and squeezed light in the LIGO detectors. The strand where quantum technology has been quietly useful for longest.
Read sheet 04 → - Sheet 05
Quantum advantage claims and the classical replies
Google's Sycamore random-circuit sampling, the Jiuzhang and Borealis boson-sampling experiments, IBM's kicked-Ising utility paper and the tensor-network simulations that answered each of them. A method for deciding how much weight a new claim deserves.
Read sheet 05 →
Notes · Method
How we weigh a claim
Quantum technology attracts announcements that are technically true and easy to misread. We do not pretend to have a formula, but we do apply the same four questions to every result before deciding how much space it deserves.
Has it been through peer review?
A preprint on arXiv is a legitimate stage of science, but it is not the same as a paper that has survived referees. We note which is which, and we treat a press release with no accompanying paper as an intention rather than a result.
What is the best classical comparison?
Every advantage claim is a race against classical computers, and classical algorithms keep improving after the quantum result is published. We ask whether the comparison used the strongest known classical method at the time, and whether anyone has since done better.
Physical or logical?
A count of physical qubits tells you little without the error rate. We try to state whether a result involves error correction, error detection with post-selection, or neither, because the three are often described with the same words.
Has anyone else reproduced it?
Independent reproduction is slow and expensive in this field, so its absence is not damning. Its presence, though, is the strongest single reason to move a claim from "reported" to "established", and we weight it accordingly.
- Logical qubit
- A qubit encoded across many physical qubits so that errors can be detected and corrected. The number of physical qubits needed depends on the code and the physical error rate.
- Threshold
- The physical error rate below which adding more qubits to a code makes the logical error smaller rather than larger. Above it, more hardware makes things worse.
- Coherence time
- How long a qubit holds its state before noise scrambles it. Useful only when read alongside gate time: what matters is how many operations fit inside it.
- Fidelity
- How close an operation comes to the ideal one. Two-qubit gate fidelity is the figure most often quoted and most often measured under favourable conditions.