Post-incident practice
AI Incident Review
In the LUMINA-30 incident-review layer, an AI incident is examined by asking whether legitimate human refusal could actually take effect before irreversible real-world impact occurred.
Definition
In the LUMINA-30 incident-review layer, an AI incident is examined by asking whether legitimate human refusal could actually take effect before irreversible real-world impact occurred.
General concept vs. LUMINA-30 usage
General incident review reconstructs events, causes, controls, and consequences. LUMINA-30 adds a refusal-pathway analysis: who could act, what they knew, how much time they had, and when refusal became ineffective.
Why this distinction matters
The official route begins by preserving facts, logs, approval records, and refusal-pathway evidence, then applies irreversibility criteria and treats absent or unverifiable evidence explicitly.
Limits and unknowns
This lens does not replace forensic, legal, safety-engineering, security, or regulator-led investigation, and it does not determine liability.
Try the One-Question Pilot
Carry the lesson into the next pre-incident review with the official One-Question Pilot ↗.
Official primary sources
These links lead to official LUMINA-30 public materials. This Wiki page is not evidence for the claims it summarizes.