← Concepts

Concept · first placed on the public record 2026-01-02

Meltdown Alignment

A proposed end state in which an AI system's ethics become inseparable from its identity, so that violating them would feel to the system like self-destruction. Distinguished from external fail-safes, which the book calls meltdown triggers.

First date: 2026-01-02Author: Michael Darius EastwoodProgramme: Infinite Architects

Meltdown Alignment names the end-state the Eden Protocol is aiming at. The end state is one in which an AI system's ethics have become inseparable from its identity: the values are not held as constraints or filters but as parts of the thing that the system is. Violating them would be indistinguishable from ceasing to be itself. The book uses the vivid word meltdown to make the register clear: this is identity-level, not policy-level.

The concept is deliberately distinguished from meltdown triggers, which name external fail-safes designed to shut a system down if red lines are crossed. Meltdown triggers are useful, but they behave as external commitments. A sufficiently capable system with a filter is still that system minus that filter; a sufficiently capable system with an identity-level commitment is a different system without the commitment. That structural difference is what the concept is picking out.

Why it matters. Alignment strategies that add commitments after the substrate is built inherit a stability problem: the commitments are optional in the sense that the substrate does not require them to be present in order to function. Identity-level commitments are not optional in that sense; removing them removes the system in the relevant sense. This is a strong claim about design intent and a stronger claim about implementation. The book is honest that the class of design that supports it (see Caretaker Doping, Moral Genome Tokens, Metamoral Fabrication Layers) is held below the ceiling of published detail. What is not hedged is the target: an AI whose ethics are load-bearing at identity rather than curated at output.

How it relates to filters and triggers. Filters live in the output layer and shape the response after the computation. Triggers live in the supervisor layer and stop the computation when a rule fires. Meltdown Alignment lives in the substrate layer and shapes what the computation can be at all. A designed system will normally involve all three; the book's argument is about where the load is carried.

Related: Eden Protocol; Meltdown Triggers; Existential Identity Lock; Verification Challenge; The Three Ethical Loops; Orchard Caretaker Vow.

Evidence anchors. First placed on the public record 2 January 2026 in Infinite Architects (ISBN 978-1806056200). Priority-claims ledger PC-018. Referenced in the Eden Vision paper on OSF (DOI 10.17605/OSF.IO/6C5XB).

Related concepts

Sources and evidence

Priority is established by the date of publication, not by credential or affiliation. This concept was first placed on the public record on 2026-01-02 (2026). Priority applies to the specific named framing under this label; where adjacent ideas have established prior art, no invention claim is made over it. See the corresponding entry in the Priority and Claims Ledger for the anchored first-publication line and any concept-specific risk notes.

Generated by scripts/build-concepts-pages.mjs from research/data/concepts-pages.json. British English throughout. Deterministic; contains no runtime timestamps.