AQ-CRED-001 — Twelve-month credibility timeline and gate mapping
PURPOSE AND STANDING
This document maps the four credibility bands defined in AQ-NODE-G0 onto programme gates AQ-GATE-04 through AQ-GATE-08, names the evidence that opens each gate, and fixes the public position permitted while a band is current. It constrains what the Akasha-Q programme may state during its first twelve funded months. It places no constraint on what may be designed or built in that period: the full reach ladder — optical bench, metro fibre and free-space, 20–30 km HAPS relay, CubeSat, operational LEO satellite, constellation — remains the programme's stated intent throughout, and the first-band restriction on claims is not a restriction on architecture.
At the time of writing nothing has been built and no element has reached laboratory validation. Every row below describes an intended design or a defined process. Present evidence level across L0–L9 is Concept, with Theory in preparation. The programme is funded from EduRankAI revenue and disciplined reinvestment; external capital, partnership and collaboration change the rate at which the bands below are traversed and never the evidence required to leave one.
THE ADVANCEMENT RULE
Evidence unlocks a level. Elapsed time never does. No gate opens on a calendar date. A gate opens when its named artefacts exist, are recorded in the Evidence & Provenance Fabric (L6), and survive review against DC-1 (every value tagged MEASURED / DERIVED / MODELLED) and DC-5 (the transition is reconstructable from the Fabric alone). Bands describe where the programme is expected to be; a band carries no authority to advance a claim.
A programme sitting at DM-2 in month 11 is in an accurate state, not a failing one. The failing state is a programme holding DM-2 evidence while making a DM-4 statement. Under DC-2 the absence of an artefact is itself a value: an unmet gate is recorded as undetermined with the missing artefact named, never as "on track" and never as a benign default.
BAND AND GATE MAP
| Band | Nominal months | Exit gate | Highest evidence level admissible | Demonstration ceiling |
|---|---|---|---|---|
| Band A — Recorded design | 0–3 | AQ-GATE-04 | Concept → Theory | DM-0 to DM-1 |
| Band B — Modelled behaviour | 3–6 | AQ-GATE-05 | Simulation | DM-2 |
| Band C — Instrumented bench | 6–10 | AQ-GATE-06, then AQ-GATE-07 | Lab validation (named quantity only) | DM-3, then DM-4 |
| Band D — Witnessed repetition | 10–12 | AQ-GATE-08 | Lab validation, repeated and observed | DM-5 |
DM-6 is not reachable inside twelve months under any evidence path, because it requires a capability repeatable outside the programme's own bench and operators. Field validation, Dynamic platform, Flight/space and Operational demonstration are levels of the Part 12 ladder that no gate in this document can confer.
WHAT UNLOCKS EACH GATE
- AQ-GATE-04 — design record admitted. L0–L9 layer definitions written as testable statements; DC-1 to DC-5 expressed so that each has a falsifying observation; trust boundaries TB-1 to TB-5 drawn on the
AQ-NODE-01block diagram with the TB-4 hardware gate identified as a physical part; conflicts CF-1..CF-7, failure modes F-1..F-14 and decision boundaries B-1..B-5 registered with a named owner and a current state. Opens: publication of architecture. - AQ-GATE-05 — model admitted. An executable model of
AQ-TSE-01covering NORMAL, RESTRICTED, CRITICAL, DENY and RECOVERING; DC-4 shown against a fault-injection set derived from F-1..F-14, with trust falling on a single observation and rising only through the full recovery sequence; command authority evaluated so that Q1–Q3 act as hard constraints and Q4 is shown, over the whole run set, only ever to reduce what Q1–Q3 allowed;abstainandsafe-holddemonstrated as reachable outcomes rather than declared ones. Opens: statements about simulated behaviour under a named assumption set. - AQ-GATE-06 — bench instrumentation admitted. L1 sensing and entropy acquisition running on real hardware inside TB-2 with deterministic timestamping; an uncertainty budget written and filed before the first measurement run; every published quantity carrying its DC-1 tag. Opens: statements about one measured quantity on one bench.
- AQ-GATE-07 — reconstruction admitted. A reviewer with no access to the operators reconstructs each consequential transition from the Evidence Fabric alone (DC-5), including at least one
denyand oneabstain; the TB-4 hardware gate shown to hold against a commanded physical effect; the correction log open to that reviewer. Opens: statements about the trust chain, not only about instruments. - AQ-GATE-08 — witnessed repetition. The Band C procedure repeated on separate days by a second operator under external observation, with dispersion reported in the quantity's own units and the assumption set restated at each run. Opens: statements that a capability is repeatable on this bench.
Figure 1 — Calendar track (descriptive) against evidence track (authoritative)
MONTH 0 3 6 10 12
|--------|---------|--------------|-----------|
BAND A B C D
(calendar position is DESCRIPTIVE - it opens nothing)
EVIDENCE TRACK (the only track that opens a gate)
[design record] --> GATE-04 --> ceiling DM-1
|
v
[executable AQ-TSE-01 model] --> GATE-05 --> ceiling DM-2
|
v
[instrumented bench, budget filed first] --> GATE-06 --> ceiling DM-3
|
v
[independent reconstruction, DC-5] --> GATE-07 --> ceiling DM-4
|
v
[second operator, separate day, observed] --> GATE-08 --> ceiling DM-5
A band may expire with its gate shut. A gate may open early.
Neither event is an error. Only a claim above the open gate is.
PUBLIC POSITION PERMITTED BY BAND
| Band | May be stated publicly | May not be stated in this band |
|---|---|---|
| A | The architecture, the ten layers, the design commitments, the reach ladder as intent, and that nothing is validated. | Any description of behaviour as observed, modelled or measured. |
| B | Behaviour observed in simulation, always with the assumption set, the model version and the fault set named in the same sentence. | Any hardware statement; any statement that a property holds outside the model's assumptions. |
| C | A named quantity measured on a named bench with its uncertainty and DC-1 tag; the reconstruction exercise and its findings. | Generalisation from bench to link, platform, or any reach rung above rung 1; any timing or ranging figure without its uncertainty. |
| D | That the procedure was repeated by a second operator on a stated date under a named observer, with dispersion reported. | Field validation, dynamic platform, flight or operational language of any kind. |
Unhackable. Impossible to intercept. Universally quantum-secure. Information-theoretically secure at system level. Satellite-ready. Space-proven. Operational. No geopolitical, sovereign-capability or defence framing in any band, at any evidence level, in any forum. Visvambhara is a separate aerospace programme and the first platform designed to carry an Akasha-Q terminal; it is never described as the purpose of Akasha-Q, and the shared record between the two programmes remains restricted-access. Deny
RECORD AND REVIEW
Each gate decision is written to the Evidence Fabric as a single entry: gate identifier, artefacts presented with their Fabric ids, reviewer, outcome from the permitted set (execute, execute-restricted, safe-hold, abstain, deny), and the specific artefact whose absence produced an abstain. Gate review may return abstain — insufficient confidence to authorize — and that outcome is filed, published to the claim register, and carries no penalty. L8 analysis may summarise a gate pack but never issues the outcome; a named human does. Bands are re-baselined at each gate opening, so a gate opened in month 4 does not compress the remaining bands and a gate unopened in month 12 rolls forward with its missing artefact still named.