Purpose
This guide provides a simple scoring aid for LUMINA-30 boundary checks.
Status
This is not a legal score, certification, or compliance rating.
It is a review aid for comparing evidence strength across cases.
Score Levels
0 — Not Applicable
The case does not involve AI-driven action, refusal authority, or irreversible impact.
1 — No Effective Refusal
Human refusal was absent, bypassed, unavailable, or operationally ineffective before irreversible impact.
2 — Not Verifiable
The review cannot verify whether human refusal was available and effective.
3 — Nominal Refusal Only
Formal human approval or oversight existed, but operational refusal effectiveness is weak or unclear.
4 — Effective Refusal Demonstrated
Records show that human refusal was available, operationally effective, and usable before irreversible impact.
Recommended Classification
| Score | Classification |
|---|---|
| 0 | Not applicable |
| 1 | Procedurally invalid for boundary-review purposes |
| 2 | Not verifiable; treat as no effective refusal |
| 3 | Weak refusal / nominal oversight |
| 4 | Effective refusal demonstrated |
Relationship to L30-CI
The 0–4 score is a review aid for comparing evidence strength.
It does not replace the final L30-CI condition label.
When a final result is recorded, use the Boundary Check Result Classification and L30-CI label:
Caution
Do not use this score to claim legal liability, regulatory compliance, or institutional certification.