← Ledger
/
Quantum Advantage Ledger · v0.2 · 2026-05-08
650 papers.
14 bills.
Empty space verified.
650-paper ledger for quantum-advantage and supremacy claims. Three signature-empty bills hold: every public claim either pays a meta-cost, is rebutted, or misses the useful-task threshold.
53/53
Classifier 1.000/1.000
Quick Orientation
Has any quantum computer actually beaten a regular one at something useful? We checked every claim.
Open brief
Every year a company announces "quantum supremacy" or "quantum advantage" on some computation. We surveyed 650 papers from 2024-2026 looking for a claim that survives a fair fight: the task is useful (not designed for the hardware), the result can be checked without trusting the quantum computer, and the classical comparison uses the best known algorithm. Every published claim either fails verification, gets matched by a clever classical algorithm within months, or relies on too few qubits to matter for real problems. The closest near-miss (Zhao-Zlokapa-Preskill-Babbush-McClean-Huang) only ran in simulation. Independent citation verification is still in progress.
Why it matters: DARPA, NIST, and every quantum-computing vendor's roadmap rest on whether 'quantum advantage' is real or just a moving goalpost.What we found: 650 papers checked. No demonstration survives all the audits. Three predicted-empty lines hold pre-verification.
Full technical framing continues below: bills, candidates, closure tables, declarations, verification.
Ledger declaration · 2026-05-08
Three signature constructions.
Six hundred fifty papers.
Empty space verified.
Bills are the closure mechanisms a quantum-advantage claim must clear. Every paper maps to one or more bills, a meta-cost, or an escape gate.
How to read this heatmap
Cells show candidate papers. A starred bill is ★ empty only when candidate count is nonzero but clean triggers remain zero after meta-cost, rebuttal, leakage, non-transfer, or escape-gate review. The closure basis appears below.
★ Predicted empty (HOLDING)
NEW v0.2 (Bill_14)
High activity (≥30 papers)
Active (10–29 papers)
★ Empty space verification
BillClosure basisCands.Clean
★ 8Cryptanalytic separation on real targets
Closure basis: M3 (asymptotic) + M5 (resource-unbounded); Wu-Vidick rebuttal closes Yilei Chen 2024 LWE in 11 dayscandidates98clean triggers0
★ 12Useful logical task >100 logical qubits
Closure basis: 5 independent multi-vendor government programs (DARPA QBI Phase 1, QuantERA EU, UK NQCC, QED-C, RIKEN/Quantinuum) all reach empty-space conclusioncandidates10clean triggers0
★ 13Heuristic + classical control
Closure basis: all 23 candidates pay M1/M3/M4/M6 or trigger Bill_13's own closure (no Pan-class problem-aware classical heuristic baseline)candidates23clean triggers0
Bill_8 (cryptanalytic separation): 98 candidates → all pay M3, M5, M4, or M6, or are declarations of empty space. Yilei Chen 2024 LWE retracted in 11 days by Wu-Vidick. Doriguello et al. $\sim 10^{13}$ qubits and $\sim 10^{31}$ years for SVP-400 even under optimistic assumptions.
Bill_12 (>100 logical-qubit useful task): 10 candidates → all pay meta-costs. 5 independent vendor-evaluation programs (DARPA QBI Phase 1, QuantERA EU, UK NQCC, QED-C, RIKEN/Quantinuum vs Fugaku) all reach the empty-space conclusion. Bill_12 emptiness is explicit policy consensus.
Bill_13 (heuristic + classical control): 23 candidates → all pay M1/M3/M4/M6 or trigger Bill_13's own closure (no Pan-class strong baseline). IBM Heron-156 optimization is the closest near-trigger.
Hardware milestones are real and substantial — but the useful-task half of Bill_12 remains absent. Industry roadmaps converge on the >100 logical-qubit threshold for 2027–2028. DARPA QBI Phase 2 Q4 2026 is the empirical crucible.
2024 Willow1 logical qubit (idle memory) · QEC scaling demo · Bill_6
2024 MS+Quantinuum4 logical qubits (tesseract code-switching primitive) · Bill_6
2024 MS+Atom24 logical qubits (logical primitives only) · Bill_6
2024 Bluvstein48 logical qubits (neutral-atom logical demo) · Bill_6
2025 Atom Phoenix50 logical qubits at 1180 atoms — largest in corpus · no useful task · Bill_6
2025 RIKEN/Quantinuum24 logical qubits vs Fugaku · no advantage
Q4 2026 — DARPA QBI Phase 23 vendors at ≥100 logical · T&E protocol operationalizes the threat model verbatim · the empirical crucible
2027–2028 IBM Kookaburra/LoonqLDPC d=12 → 100+ logical qubits (roadmap)
The most consequential single divergence in the corpus is East-West advantage status: Western/aligned programs reach empty-space consensus; CAS (Nov 2025) and USTC Hefei (Dec 2024) treat Jiuzhang 4 / Zuchongzhi 3 as advantage-demonstrated without engaging Pan-Zhang TN or Quesada-Arrazola GBS rebuttals. The ledger tracks this divergence as the most likely outcome through 2027.
GRI 2025 expert elicitation: 6% probability of CRQC by 2030. NIST IR 8528 (Sept 2025) explicitly avoids committing to a Q-Day, quietly diverging from CNSA framing within the same federal portfolio.
N1 · ★ Bill_8
Yilei Chen LWE retracted in 11 days
No 2024–2026 paper claims polynomial-time quantum cryptanalysis of a real cryptographic primitive without paying M5. Wu-Vidick `eprint:2024/583` killed `eprint:2024/555` in two weeks.
N2 · ★ Bill_12
5 independent programs find 0 of N
DARPA QBI Phase 1, QuantERA EU, UK NQCC, QED-C, RIKEN/Quantinuum vs Fugaku — all reach empty-space conclusion. Policy consensus, not framework prediction.
N3 · ★ Bill_13
No heuristic survives Pan-class baseline
All 23 candidates pay M1/M3/M4/M6 or trigger Bill_13's own closure. IBM Heron-156 optimization is the closest near-trigger; no Pan-class problem-aware classical heuristic was tested.
N4 · Bill_14 NEW
Observable estimation classically tractable
44 papers in coherent Pauli-path lineage. Boundary formalized (Rudolph-Angrisani-Holmes, Mehraban-Ghosh-Fefferman). Counter-attacks all M4-conditional.
N5
Doriguello concrete falsification
$\sim 10^{13}$ physical qubits and $\sim 10^{31}$ years for SVP-400 even under optimistic assumptions — no quantum speedup at NIST-relevant lattice dimensions.
N6
Crypto venues 6–18× faster rebuttal
Yilei Chen 11 days vs Sycamore→Pan-Zhang TN 6–18 months. Crypto venues function as peer-review filter on quant-ph rhetoric.
N7
Asiacrypt 2024+2025 zero RSA papers
Historic first — corroborated from the factorization ledger. Two flagship years with zero RSA cryptanalysis publications.
N8
QML structurally bisected
Cerezo et al. 2024: barren-plateau-free implies classically simulable. Jerbi-Dunjko-Gyurik 2026: every QML circuit on classical data with bounded fan-in admits classical surrogate.
N9
Tang dequantization extending
Quantum recommendation, qPCA, k-means, SVM, SDP — all have classical-equivalent constructions. The lineage is actively expanding.
N10
D-Wave 2024 supremacy closed
Densest rebuttal layer of any variant model: Tindall PEPS + Mauron-Wahl belief-prop + Toshiba SBM + Boixo audit + Hen-Troyer retrospective.
N11
Logical-qubit count is the new vendor metric
Vendors honestly report logical-qubit counts (Bill_6) instead of advantage-supremacy claims. Atom Phoenix at 50 logical is the largest — but no useful task.
N12
DARPA QBI Phase 1: 0 of 18 vendors
First-party government-evaluation finding aligns with our framework's empty-space prediction. Phase 2 Q4 2026 with 3 vendors at ≥100-logical is the crucible.
N13
East-West advantage divergence
Only program-level disagreement in the corpus. CAS + USTC don't engage Western classical-rebuttal lineages; every other program reaches empty-space consensus.
N14
Government Q-Day timeline divergence
NSA CNSA 2.0 + CISA aggressive 2030/2035 deadlines vs NIST IR 8528 + BSI + GRI 6% by 2030. Same federal portfolio, divergent postures.
Each negative finding becomes a checkable trigger condition. The ledger commits to public update within 7 days of any verified trigger of F8, F12, or F13.
F8 · ★ Cryptanalytic separation
Trigger: peer-reviewed quantum cryptanalytic paper that demonstrates polynomial-time attack on a real primitive (RSA, AES, NIST PQC), with concrete complexity bound, passes M5 audit, and is independently verified within 6 months
F12 · ★ >100 logical-qubit useful task
Trigger: peer-reviewed demonstration on >100 logical qubits, useful task, classical/interactive verification of output, beats best-in-class classical baseline on commodity hardware
F13 · ★ Heuristic + classical control
Trigger: heuristic-quantum advantage demonstration that beats a Pan-class problem-structure-aware classical heuristic under matched compute
F14 · Observable-estimation hardness
Trigger: non-conditional concrete construction of "observable estimation IS classically hard for circuit X" — the only territory where Bill_14 closure can be reversed
Live alerts (triggered watch-list): DARPA QBI Phase 2 Q4 2026 results (3 vendors at ≥100 logical · T&E protocol operationalizes the threat model verbatim) · Yilei-Chen NTRU 2026 successor verification · IBM Kookaburra/Loon 2027 milestone · Atom Phoenix scale-up · EUROCRYPT 2026 quantum sessions (in window).
Threat modelDemonstrate verifiable quantum advantage on a polynomial-precision task in 2026 against the best available classical algorithms running on commodity hardware (≤16 GPU, ≤TB RAM), with output that can be verified without trusting the device.
Deep loops16 sweeps × 5–10 parallel research agents per sweep × 2 batch rounds.
Sources surveyedarXiv quant-ph 2024–2026 + IACR ePrint 2024–2026 + IEEE QCE / Q2B / QIP / TQC / QSEC / AQIS proceedings + Nature, Science, PRL, PRX Quantum, npj QI, Quantum + CRYPTO / EUROCRYPT / ASIACRYPT / TCC / PKC quantum sessions + vendor announcements (Google QAI, USTC, IBM, Quantinuum, IonQ, PsiQuantum, Atom, QuEra, Microsoft) + classical-simulation rebuttal corpus (Pan-Zhang, Tindall, Bravyi-Gosset, Quesada-Arrazola, Oh-Lim, Drummond, Kalachev) + government policy (NSA, NIST, BSI, CISA, BIS, DARPA QBI, DOE QSC, EU Quantum Flagship).
ClassifierRegex rule engine. v0.2 with 53 hand-curated benchmark cases at gate-accuracy 1.000 / bill-recall 1.000.
Empty-space testThree signature bills (8, 12, 13) predeclared as empty BEFORE batch 1 sweeps. After 650 papers across 16 sweeps, all three remain empty.
ReproducibilityAll scripts, JSONs, and wiki are public. Run order: bill_classifier.py --benchmark → ledger populator → atlas review pipeline.
Every empirical claim on this page resolves to public data. Run the classifier, regenerate the heatmap, audit the corpus, file a falsification.
Locked v0.2 (May 2026) — 650 papers; Bills 8/12/13 ★ empty. Run: python3 bill_classifier.py --benchmark.
Locked state · 2026-05-08
Three signature constructions.
Six hundred fifty papers.
Empty space verified.