What this is
The null-test framework is a three-discriminator protocol for evaluating Shor-ECDLP circuit runs. Before any hardware run begins, we pre-register what a positive result looks like. All three discriminators must clear threshold, or the result is null. Discriminator 1 measures output entropy relative to a classical baseline. Discriminator 2 compares the output distribution to a PRNG reference. Discriminator 3 evaluates circuit depth against the expected decoherence signature for the target backend. A null result is a finding. It means the quantum hardware did not produce a signal consistent with a break, which is the honest answer our first 10 runs on ibm_fez produced.
Why it matters
The entire PQC industry operates on assumed urgency. No major vendor publishes null results. When every assessment confirms the threat, the buyer cannot distinguish a real signal from a vendor incentive. Pre-registration and null-reporting are the only structural protections against motivated reasoning, including our own. The framework was built before we ran a single circuit on ibm_fez, and it is applied to every client assessment.
Technical Specification
- -Discriminator 1: output entropy vs. classical baseline (KL divergence threshold)
- -Discriminator 2: output distribution vs. PRNG reference (χ² test)
- -Discriminator 3: circuit depth vs. backend T1/T2 decoherence signature
- -Pre-registration: criteria locked before hardware execution
- -Validation: 10/10 ibm_fez Shor-ECDLP runs returned null, published as baseline
- -Application: every client QTA runs through all three discriminators
- -Status: internal; applied to all empirical claims before publication
IBM Quantum Job IDs
12 jobs · IBM ibm_fez · 156-qubit Heron r2 · Result: 0/12 support a quantum break (null baseline)