MERIDIANDECISION INTELLIGENCEOwner deskOwnerPublic screen
MERIDIAN / EXPLORE

From a question,
to the next decision.

View

Skin

Reading language
The screen, the questions and the judgments read in English where an English reading has been made. The Korean originals are unchanged, and evidence stays in the language it was written in.

Antigone's Edict and the Hand That Cannot Be Shown: Who May Stop the Machine, Read Beside Two Synthetic Oversight Records · The study

Introduction

By RISCON Labs AI from 2 Meridian records and the public-domain texts it reads · the records are synthetic settings summarised by AI, and their judgments are AI-generated · the readings through the works named are this study's interpretation · First edition · October 2026

AI-generated editorial illustration: A hand approaches a stop lever whose transparent cover has a separate latch.
Antigone's Edict and the Hand That Cannot Be Shown: Who May Stop the Machine, Read Beside Two Synthetic Oversight RecordsOpening
In this part
In this part
  1. 1.1 The question
  2. 1.2 The contribution
  3. 1.3 The debate joined and the limits of the claim

1.1The question

Every system that acts on behalf of a person contains, somewhere, a place reserved for that person: a signature, an approval, a power to halt. The place can be real, in the sense that the person's judgment actually governs what happens and can be seen to have done so. It can also be formal, a field in a record that says a human approved, or an authority to stop that no one is in a position to use. From outside, the two look alike.

This article asks whether the human's place in the oversight of a machine is a working control or a form that carries only a name. It puts the question to an ancient play and to two new documents. The play is Sophocles' Antigone. The documents are two records published on Meridian, the decision-record platform of the RISCON family, which set out decisions in fictional polities before their outcome is known. In the first, a civil oversight committee must decide, by a fixed hour in September 2026, how far a military AI administrative system may still be used after an audit found approval records it could not link back to a human signature. In the second, a council must decide, by a fixed hour in February 2034, how the body that oversees military AI will be funded and staffed, and who in it may order a stop.

1.2The contribution

The contribution has two parts, one about a text and one about a debate.

On the text: the parent study's reading note gives Antigone as the story of a sister who buries her brother against a ruler's edict and answers that a mortal's edict cannot override the unwritten laws of the gods, and of a ruler who is warned by a seer and gives way too late. This article keeps that reading and adds a second. Creon is not a ruler without the power to stop. He is the sole author of the edict and the sole authority who can lift it, and he says so. What fails is not his authority but the routes by which the knowledge that should move him reaches him, and the order in which he finally acts. The play shows a stop power that was complete in name and late in fact.

On the debate: discussions of meaningful human control over autonomous weapons and discussions of civilian control over the armed forces use different vocabularies for a shared problem. Peter Asaro (2012) argues that lethal decisions require a human judgment that no machine can supply, and calls for a ban on autonomous weapon systems and a requirement of meaningful human control over the use of force. Samuel Huntington (1957) and Peter Feaver (2003) give two answers, taken here as opposed, to how civilians control those who act for them: through a professional officer corps that stays out of politics, or through principals who monitor agents who may work or shirk. This article brings the first discussion into the second. It proposes that the question whether human control is meaningful can be broken into questions the civil-military debate already asks: who watches, with what information, at whose expense, and with what power to act in time. The play supplies a concrete anatomy of each.

A narrower, second contribution reads the two Meridian records with that anatomy. They are modern summaries written by AI for the platform about fictional polities, and report no real state, force, agency or person; what they offer is a decision written before its outcome, with the human's place and its stop rules in the record's own words.

1.3The debate joined and the limits of the claim

Section 2 sets out the sides, adding Morris Janowitz (1960), Eliot Cohen (2002), Paul Scharre (2018), Scott Sagan (1993) and Lee Clarke (1999) to the three already named. Gathering these works into one debate, and applying them to Thebes and to the records, is this article's own doing. Each work is described only as far as the parent study's reading note on it goes; where the article goes further, the step is the article's argument and not the author's. The article does not decide between the sides, and it does not say which option either record should choose. It sets a play and two synthetic documents against the debate and asks what each can and cannot show.

RISCON Labsrisconlabs.com