Firebringer Quantum — Quantum Claims, Tested to Destruction

Receipts — every number independently verifiable

258 minQPU execution time on IBM quantum hardware

1,723Hardware jobs run, logged, and re-audited

$24,720Equivalent compute at IBM's published $96/min rate

2Of my own multi-year results falsified and retracted in public

Usage visible in IBM Quantum Platform analytics · job IDs published with every result

What we review

Bring us this decision.

If any of these describe your situation, the answer should be on paper before it becomes a board question.

A vendor says its product is "quantum safe." No hardware exists to test that claim. What does the evidence actually support?
A PoC team returned a quantum result. Did it survive a null test — random data through the same pipeline?
An investment memo depends on a hardware milestone. Which dependency has to fall first, and is it on track?
A board wants a defensible quantum-risk timeline. Not "5 years away" — which milestones gate the specific risk to your organization.
A partner or supplier makes a quantum advantage claim. Hardware, simulator, or emulator? What exactly was demonstrated?

Scorecard 001 · Worked example · Free

One claim, graded in the open.

The broadest claim in the market — reviewed against the five-category scorecard. Yours gets graded in private.

SCORECARD 001 · CLAIM UNDER REVIEW WORKED EXAMPLE

"Our product is quantum safe."

Hardware reality NOT DEMONSTRABLE No hardware on which this claim can be tested exists. A live "quantum resistance demo" is theater — treat it as a negative signal.

Signal integrity PARTIAL — MATH ONLY Evidence is mathematical, not empirical. Honest translation: "no known attack — yet." Proprietary PQC that skipped public cryptanalysis is near-automatic reject.

Post-processing WHERE IT DIES Implementation, key management, hybrid-mode composition, classical dependencies. Most "quantum safe" claims fail for entirely classical reasons.

Threat relevance REAL — ACT ANYWAY Harvest-now-decrypt-later is the actual risk. Federal PQC deadline is Dec 31, 2030. The threat model is sound even though the claim is untestable.

Business decision SPLIT VERDICT Reject the phrase; fund the migration. The property you can actually buy is cryptographic agility — the ability to swap algorithms when cryptanalysis moves.

Read full Scorecard 001 (PDF) See the Claims Ledger

The method

How a claim becomes a verdict.

Capture

State the claim in testable terms. Specify what it asserts, what it implies, and what decision hangs on it.

Decompose

Break the evidence structure: hardware vs. simulator, what the pipeline actually computes, what comparison is used.

Benchmark

Establish the baseline. What does a null result — or classical post-processing alone — produce?

Falsify

Run the test the claim should have faced on day one. Substitute randomized data. Apply the failure-mode checklist.

Grade

Score five categories. Each produces a finding. Together they produce a verdict: ignore, monitor, test, fund, or reject.

Translate

Convert the technical findings into the one-page board memo. Plain English. The decision and the reasoning, on paper.

Full method documentation →

The source

Justin Hughes
Founder, Firebringer Quantum

I know how convincing bad quantum evidence looks, because I built it, believed it, ran it on IBM QPUs for two years — and then killed it with the null test it should have faced on day one.

My apparent quantum signal-extraction result was reproduced perfectly by pure random noise. In parallel, a 14-bit ECDLP solve — recognized inside Project Eleven's Q-Day Prize — died under the same instrument. I retracted both results publicly. The first public falsification of a prize-confirmed submission.

Most vendor claims have never faced that test. Yours will.

1,723 hardware jobs · 258 min QPU time · 2 public self-falsifications
IBM Torino · IBM Brisbane · secp256k1 toy curves · HHL end-to-end on NISQ
Sales-to-board translation · Founder-led technical practice

IBM Quantum Platform analytics showing 258m 2s total usage and 1,723 job workloads under Justin Hughes account

IBM Quantum Platform Analytics · Account 3025655 · Justin Hughes · All time · 258m 2s · 1,723 jobs

Services

Two ways in.

Fixed scope · 10 business days

Quantum Claim Audit

Send one vendor claim, deck, white paper, NISQ result, or board question. Six deliverables. One verdict on paper — attach it to the vendor file.

$5,000 fixed · 3 slots/month

Submit a claim

Free · Ongoing intelligence

The Signal Flare

Quantum claims intelligence for security, risk, strategy, and investment decision-makers. Milestone tracking, announcement translation, dependency mapping — not news coverage.

Monthly · No card required

Join the Signal Flare

Incentives, disclosed

We sell analysis and nothing else. No migration products, no hardware, no tokens, no vendor referral fees. We do not sell quantum optimism or quantum cynicism. We sell quantum defensibility. A verdict of "reject" costs us nothing and a verdict of "fund" earns us nothing — which is the only condition under which either verdict is worth paying for.

Every claim we publish carries a status in the open Claims Ledger. Every result we have produced — including two that we retracted — is documented in Research & Corrections. The instrument has no favorites, including us.