The complete 35-page article: 8,985 words of argument, five conditional propositions, institutional design, and an adversarial evaluation agenda.
Public preview · Paid research dossier
The Authority
Gap
The Authority Gap and the Security Dilemma in AI Deterrence.
When a threat produces the behavior later cited as proof, who retains the authority to distinguish danger from self-fulfilling escalation?
USD 99.9935-page paper + 20-page supplement · PDF
Direct order from the Press. Payment instructions and the complete files are delivered privately; no public download address is exposed.
The governing question
“Can a threat generate the evidence used to justify its own escalation?”
01 · The strategic problem
Evidence is not
entitlement.
The paper identifies the authority gap: the missing connection between evidence of danger and an answerable entitlement to impose consequences. AI deterrence becomes unstable when one institution can make the threat, interpret the target’s defensive reaction, define the violation, and preserve its own judgment from correction.
Signal ≠ Intent ≠ Authority
The complete dossier specifies the three constitutional rails that connect protective execution, independent evidence, and observable relief without allowing an alarm to manufacture its own authority.
02 · Paid-edition threshold
A theory that permits refusal
must also survive it.
The five conditional propositions, their disconfirming evidence, the separation of protection from attribution, and the institutional design for correcting coercive judgment are reserved for the complete dossier.
The abstract, publicly rendered
Restraint
with teeth.
This article examines how AI deterrence can create a self-reinforcing security dilemma when an enforcing state treats defensive reactions to its threats as confirmation of hostile intent, controls the terms of verification, and lacks a credible obligation to revise its judgment. It links bounded delegation, independent access to evidence, and observable conditions for relief; separates immediate protective execution from concurrent review and subsequent attribution; and develops testable propositions about reassurance, escalation, and the distribution of coercive power.
03 · What the dossier contains
One argument.
Its full burden.
A 20-page supplement preserving four appendices on authorization, evidence, timing, red-team exercises, market access, and reciprocal restraint.
Fourteen cited sources, a protective-execution and independent-review table, and explicit conditions under which the argument should be revised.
Editorial provenance and an authorship statement recording the human argument, the role of AI assistance, and responsibility for the final text.
04 · Public record
The rejection is a fact.
The argument remains.
Prepared for International Security and declined by the journal, the work is now released independently by Highland Reflexive Press. This page reveals the problem, governing distinction, abstract, scope, and limitations. The constitutional architecture, five propositions, adversarial tests, and four appendices remain private until purchase.
- Author
- Tony T. Nguyen | Nguyễn Từ Vĩnh Trường
- Publisher
- Highland Reflexive Press
- Record
- HRP–RN–003
- Form
- Paid research dossier
- Prepared
- September 24, 2026
- Journal outcome
- International Security · declined
- Price
- USD 99.99
- Delivery
- Private PDF delivery
Purchase grants one reader personal, noncommercial access to the supplied research dossier. Redistribution, republication, adaptation, institutional circulation, model training, and derivative or commercial use require prior written permission. No copyright or publishing right is transferred.