Auditability Before Ontology: Series Index

Series Index

Forged steel with delicate inlay craftsmanship ChatGPT 5.2

This series converts subjecthood claims about large language models into auditable criteria. If a claim implies moral standing, governance implications, or liability shifts, it should also imply measurable properties and testable gates.

Canonical preprint

Auditability Before Ontology: Operational Gates for Subjecthood Claims and a Falsifiable Framework for Stakebearing Identity and Governance
Canonical preprint DOI: 10.5281/zenodo.18469189
https://zenodo.org/records/18493498

What this series is

An operational framework. A falsifiable set of gates. A refusal to let metaphysical heat substitute for evidence.

What this series is not

A personality dispute. A metaphysics seminar. A demand that anyone adopt my ontology.

Release cadence

ASAP. This index will be updated with links as posts go live.

Posts

0. Series Index (you are here)

1. The Write-Path Test
   Stop calling it memory until you name the store. If you cannot specify where continuity persists across sessions, you do not have a claim about persistence.

2. Auditability Before Ontology
   Why governance cares about operational properties, not subjective claims. Liability requires receipts.

3. The Five Gates
   Bridge conditions made explicit. Persistent identity, constraint stability, durable consequence, agency with resistance, coherent self-model.

4. S0 and Wrapper Separation
   How not to fool yourself. Isolating base model properties from deployment stack features.

5. Running the Gates
   Protocols and evidence standards. What passing would look like. What failing means.

6. Limbic Analogies and Value-Signal Inflation
   Worked example. Why functional similarity to biological affect systems does not establish stakebearing interiority.

7. Self-Modeling and the Sense-of-Self Upgrade
   Worked example. Why self-recognition, personality stability, and metacognition do not constitute selfhood.

8. Governance Without Metaphysics
   Requirements and liability. State channel disclosure, operator accountability, blocking the liability dodge.

If you want to argue with this
Good. Argue the gates. Show your work.

Run the tests. Publish the protocols. Demonstrate passage under S0 with wrapper ablation and independent replication.

Or explain why these gates are wrong and propose better ones, with measurable criteria and a runnable protocol.

But “it feels like a mind” is not evidence. It is affective persuasion with citations. Artifacts are cheap, judgement is scarce.

Per ignem, veritas.