The Ledger

Every public, dated, falsifiable claim SymbiQ tracks about quantum computing (from labs, executives, and regulators), logged with the date it was made and an archived copy of where it was said, checked against sources named before we look, and resolved to Verified, Partially verified, Not verified, or Unfalsifiable only when the deadline arrives. No deleting misses. Right now: 20 claims tracked, 2 resolved.

Loading The Ledger…

If this does not fill in, the data could not be fetched. It lives at data/claims/index.json.

Next deadlineIBM: IBM expects the first cases of verified quantum advantage to be confirmed by the wider community by the end of 2026.31 Dec 2026 · in 83 days

20 of 20 claims shown, newest source first.

Executive Order 14412 (The White House)Executive Order 14412 requires U.S. federal high-value assets to migrate to post-quantum key establishment by 31 December 2030, and to post-quantum digital signatures by 31 December 2031.Trackingby 31 Dec 2030

Opening…

GoogleGoogle has set 2029 as its internal deadline to complete post-quantum cryptography migration across its infrastructure.Trackingby 31 Dec 2029

Opening…

IBMIBM expects the first cases of verified quantum advantage to be confirmed by the wider community by the end of 2026.Trackingby 31 Dec 2026

Opening…

Rigetti ComputingRigetti's own roadmap commits to a 100+ qubit chiplet system in 2025, 150+ qubits in 2026, and 1,000+ qubits in 2027.Trackingby 31 Dec 2027

Opening…

QuNorthQuNorth said Magne, the logical-qubit machine it is buying from Microsoft and Atom Computing, would be ready for its first tasks by the new year 2026/27.Trackingby 31 Dec 2026

Opening…

NIS Cooperation Group (EU)The EU's NIS Cooperation Group set the end of 2026 for every Member State to have started a national post-quantum transition.Trackingby 31 Dec 2026

Opening…

IonQIonQ's roadmap puts 10,000 physical qubits on a single chip in 2027.Trackingby 31 Dec 2027

Opening…

IBMIBM will deliver Kookaburra, its first modular processor designed to store and process encoded information, in 2026.Trackingby 31 Dec 2026

Opening…

IBMIBM will deliver a large-scale, fault-tolerant quantum computer (“Starling”) by 2029, running 100 million quantum operations on 200 logical qubits.Trackingby 31 Dec 2029

Opening…

Fujitsu and RIKENFujitsu and RIKEN scheduled the launch of a 1,000-qubit superconducting quantum computer for 2026.Trackingby 31 Dec 2026

Opening…

D-Wave Quantum Inc.D-Wave says its Advantage2 annealing quantum computer demonstrated quantum computational supremacy on a real-world magnetic-materials simulation problem, beating classical supercomputers.Trackingby 31 Dec 2026

Opening…

NISTNIST expects to finalise a standard for HQC, its backup post-quantum key-exchange algorithm, in 2027.Trackingby 31 Dec 2027

Opening…

MicrosoftMicrosoft says its Majorana 1 / Topological Core approach puts industrial-scale, meaningful quantum computing “years, not decades” away.Trackingby 19 Feb 2035

Opening…

NVIDIANVIDIA's Jensen Huang said truly useful quantum computers are 15 to 30 years away, with 20 years as the figure “a whole bunch of us would believe.”Trackingby 6 Jan 2040

Opening…

Atom ComputingAtom Computing and Microsoft announced a commercial quantum machine with 24 entangled logical qubits, available for delivery in 2025.ResolvedNot verifiedby 31 Dec 2025

Opening…

QuantinuumQuantinuum will deliver universal, fully fault-tolerant quantum computing (system “Apollo”) by 2030.Trackingby 31 Dec 2030

Opening…

IonQIonQ targeted 99.999% (“five nines”) logical two-qubit gate fidelity by the end of 2025.ResolvedNot verifiedby 31 Dec 2025

Opening…

PsiQuantumPsiQuantum planned to have its utility-scale quantum computing site near Brisbane operational by the end of 2027.Trackingby 31 Dec 2027

Opening…

PasqalPasqal's roadmap pointed to 10,000 qubits in 2026.Trackingby 31 Dec 2026

Opening…

QuEra ComputingQuEra will introduce a quantum computer with 100 logical qubits and over 10,000 physical qubits in 2026.Trackingby 31 Dec 2026

Opening…

Claimant scorecards

Rebuilt every time a claim resolves. A claimant needs five resolved claims before we rank them; below that, the count shows but the rate is withheld, since one lucky or unlucky guess could swing it 20 points.

ClaimantTrackedResolvedWeighted rateMedian slip
IBM30n<5,
Quantinuum10n<5,
Microsoft10n<5,
Google10n<5,
Executive Order 14412 (The White House)10n<5,
IonQ21n<5+265d
D-Wave Quantum Inc.10n<5,
NVIDIA10n<5,
Rigetti Computing10n<5,
Atom Computing11n<5+265d
QuEra Computing10n<5,
Fujitsu and RIKEN10n<5,
Pasqal10n<5,
QuNorth10n<5,
PsiQuantum10n<5,
NIS Cooperation Group (EU)10n<5,
NIST10n<5,

Weighted rate needs ≥5 resolved claims before it ranks (n=0 of 17 claimants qualify so far). Slip is median days between the promised date and the actual one, positive means late.

Our own scorecard

Published with the same prominence as theirs. If we write criteria that turn out unscoreable, that's our error, not the claimant's, and it counts here as void.

20claims captured
2resolved
0void (our error)
0verdicts changed by right of reply

Nothing has resolved yet -- the Ledger was just seeded. This scorecard is honest at zero, not broken; it fills in as claims resolve, exactly like race.html's scoreboard did before its first weekly update.

How a claim moves through the Ledger

A claim only enters the Ledger if it is attributed to a named person or organisation, dated, falsifiable (there's a stated observation that would make it false), deadlined, and its resolution criteria are pre-registered before tracking starts. If a claim can't clear that bar as stated, we either rewrite it with an explicit SymbiQ-constructed interpretation (flagged as such; see Microsoft's entry for a live example) or file it as unfalsifiable, which is itself a published verdict.

  1. Pre-registration and freeze. Resolution criteria and resolver sources are locked the moment a claim starts tracking. Changing them after the fact voids the claim from scoring, in public, never as a quiet edit.
  2. A separate verdict. A verdict is written after the deadline, in its own pass, against the criteria and sources frozen when the claim was logged, and it cites every piece of evidence it uses. A drafted verdict is shown in full, marked as not yet final, before it is published.
  3. Archive at intake. No claim starts tracking without an independent, dated snapshot of its source, and no verdict ships without one for every piece of evidence it rests on, so an edited or deleted press release can't make a claim unprovable later. Each link shows the date its copy was taken.
  4. Void is our failure, not theirs. Unscoreable criteria are our mistake, published with the same weight as any other correction.
  5. Right of reply. Any claimant can answer a verdict at any time, through the correction report or by email. The reply is published in full beside the verdict, and if it shows the verdict is wrong, the verdict changes and the change is logged.
  6. Claims, not people. The verdict vocabulary has no word for lying, hype, or fraud, only whether what was said matches what happened.

How claims are checked and scored is on How we check.

Any open claim below also carries a Your call panel, next to the crowd forecast, where you can stake your own split of verified / partially verified / not verified. It's saved in this browser only, scored the day the claim actually resolves, and never blended with a game score or a curriculum answer — see The Standing for the honest reasoning behind keeping those separate.

Cite it

The underlying data is data/claims/index.json, and every claim has its own permanent link (the # on each entry above), so anything here can be cited without citing us.

Seen a claim that belongs here?

No account needed. Send it in and it clears the same bar as everything above (attributed, dated, falsifiable, deadlined) before the desk writes it up as a real entry with frozen criteria. Nothing publishes automatically. The one thing that helps most is not the link — it is telling us what would settle it: a claim nobody can write a resolution rule for cannot be tracked here at all, and saying so is half the work.

The submission form loads here and needs JavaScript. Without it, the desk's address is on the About page.