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.

Four Reports, Three Sources: What a Count of Original Sources Decides in Ten AEGIS Training Cases · The study

Sources and bibliography

By RISCON Labs AI from AEGIS's public records · the cases are authored synthetic training cases, not real events, and their reports, offices and figures are the author's assumptions; their English is this study's translation of the Korean records · the readings through the works named are this study's interpretation · First edition · October 2026

AI-generated editorial illustration: Leaf impressions beside a carved block and a second block turned over.
Four Reports, Three Sources: What a Count of Original Sources Decides in Ten AEGIS Training CasesOpening
In this part
In this part
  1. Bibliography

This study is an expansion by AI of ten decision records published by AEGIS, a studio in the RISCON family. Every case is an authored synthetic setting: AEGIS marks each one realistic synthetic, calls each record an authored question seed whose judgment mark is the author's record and not a factual basis from public records, says of the programme that every number is a value assumed by the author and has not been compared with public records, and says in each run that the mapping of reports to original sources is the author's assumed value. The reports, offices, ledgers and documents the cases speak of are therefore part of the authored setting, not sources this study or AEGIS cites. AEGIS published the records in Korean; the English in this study is a translation made once by the editorial session of this study, sentence by sentence, and each case's English seed names the Korean original by its hash. Every sentence in a block quote is that English, word for word; the rest is AI writing. No case carries a completed decision: each carries the author's marked recommendation and a small executed calculation, and waits for the rest of its input. The readings through the works named are this study's interpretation, and it quotes nothing from outside the records.

CaseStudioRecord
Maintenance history reportsAEGIShttps://aegis.risconlabs.com/#issues?case=AEG-K-021
Three kinds of personnel reportAEGIShttps://aegis.risconlabs.com/#issues?case=AEG-K-039
Press releases from three placesAEGIShttps://aegis.risconlabs.com/#issues?case=AEG-K-096
Maintenance performance reportsAEGIShttps://aegis.risconlabs.com/#issues?case=AEG-K-112
Fuel consumption reportsAEGIShttps://aegis.risconlabs.com/#issues?case=AEG-K-113
Training accident reportsAEGIShttps://aegis.risconlabs.com/#issues?case=AEG-K-147
Supply statisticsAEGIShttps://aegis.risconlabs.com/#issues?case=AEG-K-172
Scattered situation reportsAEGIShttps://aegis.risconlabs.com/#issues?case=AEG-K-177
Records of historical decisionsAEGIShttps://aegis.risconlabs.com/#issues?case=AEG-K-178
An alliance consultation's documentsAEGIShttps://aegis.risconlabs.com/#issues?case=AEG-K-179

The records were read on 2026-10-06. This book was written on 2026-10-07.

Bibliography

  • Martin Landau, Redundancy, Rationality, and the Problem of Duplication and Overlap (1969).
  • A. Avizienis, The N-Version Approach to Fault-Tolerant Software (1985).
  • D. E. Eckhardt and L. D. Lee, A Theoretical Basis for the Analysis of Multiversion Software Subject to Coincident Errors (1985).
  • John C. Knight and Nancy G. Leveson, An Experimental Evaluation of the Assumption of Independence in Multiversion Programming (1986).
  • Krishna K. Ladha, The Condorcet Jury Theorem, Free Speech, and Correlated Votes (1992).
  • Alvin I. Goldman, Experts: Which Ones Should You Trust? (2002).
  • Luc Bovens and Stephan Hartmann, Bayesian Epistemology (2004).
  • Michael Weisberg, Robustness Analysis (2006).
  • Martha S. Feldman and James G. March, Information in Organizations as Signal and Symbol (1981).

Every sentence in a quotation block is AEGIS's record, word for word. The rest is AI's writing from those records and the works the book names, and it does not say which choice is right.

RISCON Labsrisconlabs.com