L30-RET — The Refusal Option and LUMINA-30

Research-to-practice guide · public research layer · non-canonical

One sentence: The Refusal Option asks whether an independent Human NO can still change the outcome within the remaining time before irreversibility, and provides L30-RET — the LUMINA-30 Refusal Effectiveness Test (RET) — to test whether refusal is operationally meaningful.
Part of the wider Public Research Stack. L30-RET is the human-refusal layer alongside PCR-C, Objective Persistence, and PISF. Explore all four research lines.

Relationship to LUMINA-30

LUMINA-30 asks whether accountable humans can still effectively refuse, stop, verify, or redirect before AI-related consequences become irreversible. The Refusal Option deepens that question at the human-standing and decision-governance layer. It supports the boundary question; it does not replace or redefine the LUMINA-30 canon.

L30-RET — Refusal Effectiveness Test (RET)

L30-RET is the LUMINA-30-specific public discovery identifier for the existing RET framework. It does not introduce a new theory, certification, or safety guarantee.

L30-RET (RET) is a non-compensatory seven-gate framework. Human presence, formal approval, or a nominal stop control is not sufficient by itself.

1. Authority
2. Decision-Relevant Information
3. Practical Access
4. Temporal Sufficiency
5. Causal Efficacy
6. Execution Independence
7. Epistemic Autonomy

L30-RET (RET) is time- and scope-dependent. A positive result at one state does not create permanent certification.

Timing, causality, and change

The research separates total human intervention latency from the remaining interval to irreversibility. It also asks whether do(NO) would actually change the relevant outcome, and treats a positive RET result at one time as insufficient to establish the same result after material change.

Public research roles

LayerRole
Paper 1 — PCR-CPre-irreversibility infrastructure/control research.
Paper 2 — Objective PersistenceExistence-condition / external-anchor supporting research.
Paper 5 — L30-RET / The Refusal OptionHuman-standing / effective-refusal / decision-governance research.
Paper 6 — PISFIntegrated authority / refusal / human-revisable-development research.
LUMINA-30Public boundary framework and practical reference.

Research supports LUMINA-30; research does not define the canon.

Paper numbering

Paper numbers follow internal research chronology, not publication sequence. The current public research set contains Papers 1, 2, 5, and 6. Papers 3 and 4 are internal unpublished research and are outside the current public LUMINA-30 research set. Their content should not be inferred, reconstructed, or cited from the numbering.

Use with existing LUMINA-30 tools

The following is a routing guide, not an official certification sequence or mandatory compliance workflow.

  1. Start with the LUMINA-30 boundary question and identify the potentially irreversible transition.
  2. Use L30-RET (RET) when oversight may be only formal.
  3. Route concrete gaps to existing tools: Stop Authority, Evidence Requirements, PCR-C, Institutional Friction, Accountability, One-Question Pilot, Boundary Check, L30-BAS, and L30_FRM.
  4. Preserve decision records and evidence.
  5. Revalidate after material change.
  6. Use Incident Review after failure, near miss, or disputed control.

Limits

The paper does not claim absolute invulnerability against superintelligence, a containment theorem, permanent Human supremacy, a universal veto, LUMINA-30 certification, a safety guarantee, legal approval, or peer-reviewed status merely from Zenodo publication.

Stable references