How a Quantum Computer Works
A quantum computer replaces the ordinary bit, a switch that is either zero or one, with a qubit: something that holds both at once until it is measured. This video explains what that actually means, why building one is so hard, and where the technology genuinely stands today.
First the mechanism: superposition, measurement, and entanglement, explained without the shortcuts that make the subject sound more mysterious than it is. Then the physical machine - superconducting qubits cooled near absolute zero, and decoherence, the fragility that has limited the field since it began.
Then the milestone: Google's Willow chip, reported in Nature on 9 December 2024, was the first real device where adding more physical qubits made the error rate go down rather than up - 'below the surface code threshold'. Its logical qubit lasted 2.4 times longer than its best individual physical qubit.
Then the plan: IBM's roadmap, published 10 June 2025, targets a machine called Starling for 2029 - 200 logical qubits, 100 million quantum operations - and a later machine, Blue Jay, with no date given.
And the contested claim: Microsoft's Majorana 1 chip, announced February 2025, claims a sturdier kind of qubit. Nature published a formal challenge to the evidence, echoing a near-identical claim from 2018 that was later retracted. This video reports it as contested, not settled - neither side has closed the question.
Finally, what any of this is actually for: a narrow set of problems with a known quantum algorithm, including factoring the large numbers behind today's internet encryption - the subject of this channel's companion video, 'The Quantum Threat to the Internet', which this video does not repeat.
No figure appears without its date, unit and source. A contested claim is presented as contested.
Educational documentary. Not financial or investment advice.
In these topics
Tags
- quantum computing explained
- how quantum computers work
- qubit explained
- superposition
- quantum entanglement
- Google Willow chip
- quantum error correction
- IBM quantum computing
- IBM Starling
- Microsoft Majorana 1
- topological qubit
- decoherence
- quantum threshold
- quantum supremacy
- science explained
Sources and credits
Original narration, artwork, score and editing assembled locally for Forwards or Backwards. No third-party footage, music or imagery.
- Narration: Kokoro-82M
bf_emma, British English. - Imagery: generated locally, 10 environments, 10 matched states.
- Score: original procedural pad, ducked under the narration.
Primary sources
- Google Quantum AI and collaborators, 'Quantum error correction below the surface code threshold', Nature, published online 9 December 2024 (nature.com/articles/s41586-024-08449-y) - Willow chip; 72-qubit (distance-5) and 105-qubit (distance-7, 101-qubit code) processors; error suppression factor Lambda = 2.14 +/- 0.02 per code-distance step; logical error rate 0.143% +/- 0.003% per cycle at distance 7; logical qubit lifetime 2.4 +/- 0.3 times its best physical qubit. READ IN FULL 22 September 2026.
- Google, same-day announcement (9 December 2024) - benchmark sampling task, about five minutes on Willow, company estimate of about 10 septillion (10^25) years on a leading classical supercomputer for the same task. Reported as Google's own estimate for one benchmark, not an independent measurement.
- IBM Newsroom, 'IBM Sets the Course to Build World's First Large-Scale, Fault-Tolerant Quantum Computer at New IBM Quantum Data Center', newsroom.ibm.com, 10 June 2025 - IBM Quantum Starling targeted for 2029 at a new IBM Quantum Data Center in Poughkeepsie, New York; 200 logical qubits; 100 million quantum operations; qLDPC codes cutting physical-qubit overhead by about 90% against the surface-code approach; IBM Quantum Blue Jay named as the following machine (2,000 logical qubits, 1 billion quantum operations), no date given.
- Microsoft, Majorana 1 announcement, February 2025 - described by Microsoft as the first quantum processor built on a 'Topological Core'; eight qubits.
- Nature, published challenge to the Majorana 1 claim (nature.com/articles/d41586-025-00683-2) - the published measurements do not, by themselves, distinguish a genuine Majorana zero mode from an ordinary Andreev state.
- Background, cross-checked: a Microsoft-affiliated Nature paper (2018) claiming conclusive evidence of Majorana zero modes was later retracted after the same data proved consistent with Andreev states.
- NOT USED: any claim that a working general-purpose quantum computer exists today; any resolution of the Majorana dispute; parameter or cost figures beyond what each company has itself published and dated.
Figures were not re-pulled within 48 hours of export; re-verify before publication if the film is held.
Not regulated financial advice.