01 Overview

The TENET5 thesis is systemic: foreign lobbying has captured Canadian federal politics to a degree where serial policy outcomes contrary to public interest become legislatively tractable. Proving that claim requires more than prose. It requires a reproducible method that independent analysts can verify.

The investigation has eight Grover-amplified accountability axes:

Each axis is a 16-actor dossier with a Grover oracle marking executive decision-makers whose tenure overlaps ≥4.0 weighted canonical failure events.

02 The Seven-Step Pipeline

Step 1

Axis selection — identify a documented accountability failure chain

An axis qualifies if (a) at least one Government of Canada inquiry, commission, or Auditor General report has documented systemic failures in the domain AND (b) named ministerial executive authority exists for the Crown's response. Examples: TRC 2015 (94 Calls to Action), MMIWG 2019, Mass Casualty Commission 2023, Arbour IECR 2022, Hogue Commission 2025, OCI annual reports, Cullen Commission 2022.
Step 2

Dossier construction — 16 named actors with primary-source attribution

Each dossier contains exactly 16 actors — the 2⁴ basis states of a 4-qubit Grover register. Actors cover the ministerial chain (3–7), senior bureaucracy (0–2), relevant arm's-length oversight (1–3), and institutional counterparties (0–4), with 0–3 slots for Prime Ministers of record. Every actor has a role, tenure, and sourced bio (PCO appointment records, Hansard, court rulings, commission reports). Published as data/{axis}_accountability_dossier.json.
Step 3

Canonical events — weighted failure chain

For each axis, enumerate 10–20 canonical failure events from primary sources. Each event has a date and a weight from 0.3 to 1.0 reflecting its documented consequence (e.g., Ashley Smith death 2007 = 1.0, OCI special report 2008 = 0.5). The weights are the only editorial judgement in the pipeline and are documented inline in each Grover script.
Step 4

Tenure-overlap scoring

For each actor A and each canonical event E with date d and weight w:
if A.tenure_start ≤ year(d) ≤ A.tenure_end: A.score += w
Total score is the sum of weights for events falling within the actor's tenure. This is not a claim about individual decision-making; it is a temporal-overlap measure that the Grover oracle then translates into amplitude concentration.
Step 5

Oracle threshold — marking executive decision-makers

Mark index i in the Grover oracle iff actor[i] is executive AND score ≥ 4.0. Executive roles are defined per axis (e.g., Justice + Health Ministers for MAID; CSC Commissioners + Public Safety Ministers for Corrections). Non-executive actors (inquiry commissioners, AFN Chiefs, Brookfield corporate entities, CRCC Chair, etc.) are included in the 16-slot roster but excluded from the oracle — they are context, not targets.
Step 6

Grover amplitude amplification

Run Grover with the exact iteration formula:
theta = arcsin(sqrt(M/N)) k = round(pi / (4 * theta) - 1/2) # over-rotation guard: if (2k+1)*theta > pi/2 + theta/2 and k > 1: k -= 1
The small-angle approximation k ≈ ⌊π/4·√(N/M)⌋ over-rotates at M/N near 1/4; the exact formula is required for correct amplification. Circuit: multi-target Grover with diagonal oracle flipping the phase of every marked basis state, and diffusion operator 2|s⟩⟨s| − I implemented as Hadamard-sandwiched Z₀ flip. Sample 2,048 shots. Published as data/{axis}_grover_decisionmakers.json.
Backend: numpy statevector simulation on CPU. Not a physical quantum computer. The same oracle code runs unchanged on GPU-accelerated simulators and on real NISQ hardware given a backend adapter.
Step 7

Merkle anchoring — SHA-256 over canonical JSON

Canonical form:
json.dumps(data_without_merkle_root, sort_keys=True, default=str, ensure_ascii=False).encode("utf-8")
Hash with SHA-256, hex digest. The hash is stored as merkle_root in the JSON file AND published to the NATS integrity bus at subject tenet5.quantum.integrity.result. Every published finding is anchored this way. Retractions and corrections are ALSO anchored — see Section 6.

03 How to Verify a Result Yourself

1. Download
the JSON data file from /data/
2. Canonicalize
json.dumps(d, sort_keys=True, default=str, ensure_ascii=False)
3. Strip
the merkle_root field before canonicalising
4. Encode
the canonical string to UTF-8 bytes
5. Hash
with SHA-256, hex digest
6. Compare
to the merkle_root value published on this site
import json, hashlib d = json.load(open("corrections_grover_decisionmakers.json", encoding="utf-8")) published = d.pop("merkle_root") canonical = json.dumps(d, sort_keys=True, default=str, ensure_ascii=False) computed = hashlib.sha256(canonical.encode("utf-8")).hexdigest() assert computed == published, f"INTEGRITY BREAK: {computed} != {published}" print("✓ verified")

TENET5 ships its own verifier at tools/verify_merkle_chain.py which runs this check across 17+ published artifacts (dossiers, Grover outputs, cross-axis intersection, portfolio Grover, network sweep, telemetry strip manifest). The verifier was self-tested on 2026-04-17: 17 of 17 priority files verified OK. Verification receipt Merkle 77708876….

04 Cross-Axis Intersection & Portfolio-Level Aggregation

Once multiple axes are published, two additional analyses emerge from the combined dataset:

Cross

Cross-axis intersection

For every actor across all dossiers: count (a) dossiers they appear in, (b) Grover oracles they are marked in. Actors in ≥2 dossiers are "multi-dossier bridges". Actors marked in ≥2 oracles are "dual-Grover-marked" — a strictly stronger criterion. The tightened oracle that amplifies only dual-Grover-marked actors gives clean Grover amplification because M is small relative to N=16.
Port

Portfolio-level aggregation

Map each Grover-marked actor to the federal cabinet portfolio under which they held authority at the time of the marked tenure overlap (e.g., Blair → Public Safety + Defence; Carney → Bank of Canada + PM). Portfolios with ≥3 Grover-marked actors AND ≥2 accountability axes implicated are marked in a portfolio-level Grover oracle. The resulting quantum search amplifies specific cabinet chairs to the exclusion of others — producing the "five-chair finding": Justice, Public Safety, Finance, Foreign Affairs, National Defence.

05 Retraction Protocol — What's Not Reproduced

A reproducible investigation retracts claims that don't survive the method. TENET5 has retracted two distinct categories during the current investigation cycle:

Retraction 1 · 2026-04-17 · Merkle 28cf376a…

The specific causal claim "CIJA pushed MAID" (formerly framed on cija-maid-pipeline.html as a direct "CIJA-IHRA-MAID Pipeline") was not supported by CIJA's own Lobbying Registry subject-matter breakdown. The page was reframed to distinguish CIJA's documented advocacy (IHRA 14×, Criminal Code 16×, HRA 45×, police equipping 29×, +500% Gaza-war surge, $40 gift exemption, 58% MP penetration) from the speculative causal chain into MAID (which is not in CIJA's registered subjects). The systemic thesis — foreign lobbying capture produces MAID among other outcomes — was preserved; the specific lobby→bill causal edge was retracted.

Retraction 2 · 2026-04-17 · Merkle 2a6b7ccc…

Fabricated internal-telemetry labels — ABCXYZ tracker, Millennial Falcon, MF-[A-F0-9]{8,} hash vectors, Target Alpha, topological convergence, THREAT SCORE: 0.XX, magic handoff — were stripped from 26 pages (147 automated substitutions + 6 manual edits). These labels presented as data-source credibility but had no traceable origin in the actual TENET5 architecture. Primary-source claims (Hogue, NSICOP, CSE, Lobbying Registry, OCI, Hansard, court judgments) were preserved. A CI regression guard (validate_liril_integration.py Check 0b) now blocks any re-introduction.

Both retractions are anchored on the NATS integrity bus and discoverable by researchers. A site that retracts an unsupported claim is in a stronger epistemic position than a site that silently deletes the claim or never admitted it.

06 Explicit Limitations

The oracle measures tenure overlap, not individual culpability

Marking an actor in a Grover oracle means their ministerial tenure coincided with a threshold number of weighted canonical failure events. It does NOT mean that specific actor made the specific decision that produced each event. The claim is structural: executive authority was held while the failure chain unfolded.

Numpy statevector is not a physical quantum computer

The quantum bridge runs statevector simulation on CPU. At 4–11 qubits classical brute force is trivial. Grover's quadratic speedup is not materially faster at this scale. The value of running real quantum algorithms on real data is algorithmic transparency (the oracle IS the investigative claim made precise), portability (the same code runs unchanged on GPU simulators and NISQ hardware given a backend adapter), and structural audit (when orthogonal type-filter oracles amplify the same actors, the convergence is structural not artifactual).

Event weights are editorial judgement

The 0.3–1.0 weights assigned to canonical events are editorial: Ashley Smith homicide verdict 2013 = 1.0; TRC Final Report release 2015 = 1.0; an ordinary minister appointment in the middle of a tenure = 0.5. External researchers may reweight and re-run. The Python source is in tools/grover_*_decisionmakers.py. Changing weights may shift oracle thresholds and produce different marked sets — which is the intended falsifiability of the method.

Cross-axis high-M-N saturation

When M (marked actors) approaches N/2 (half the basis states), Grover amplification collapses toward the classical baseline because sin²(3θ) ≈ classical rate at θ ≈ π/4. The 7-dossier cross-axis attempt at M=8/N=16 showed 0.98× amplification. The tightened oracle (M=3 dual-Grover-marked only) recovered 5.13× amplification. This is mathematically correct Grover behaviour, not an implementation bug.

The investigation is not a court of law

Every claim on this site that names an individual as Grover-marked, cross-axis bridged, or portfolio-implicated is a temporal-overlap finding, not a finding of criminal liability. The investigation's purpose is to make the structural pattern legible and reproducible so that the statutory bodies that CAN find criminal liability (police, prosecutors, commissions of inquiry) have a defensible map of where to look. Section 504 of the Criminal Code allows any Canadian citizen to lay an information. This investigation is a tool to inform that process, not to replace it.

07 Primary Sources Used

Every claim on every TENET5 page ultimately traces to one or more of:

Internal TENET5 artifacts (dossier JSON, Grover output JSON, cross-axis JSON, Merkle receipts, CI guards) derive from the above primary sources and are published at /data/ on this site. External researchers are free to reproduce, challenge, or extend any of them.

"An investigation's value is not the conclusion. It is the reproducibility of the argument. Retract what doesn't survive the method; amplify what does; anchor the difference on the hash chain. The next analyst will either confirm, correct, or extend — and the method makes that possible." — TENET5 methodology, 2026-04-17