OLYMPUS RISK INTELLIGENCE PROTOCOL — HUMAN THREAT ASSESSMENT DIVISION

IAN GORRIE

CASE: OLYMPUS-000 (the cataloguer)
STATUS: ACTIVE — Author. Self-published. Answers to no institution, which the protocol notes as both a finding and a grievance.
THE WITNESS — DOCUMENTS THE APPARATUS FROM INSIDE IT
23
HAZARD SCORE

Behavioral Archetype

THE WITNESS — Subject compiled every other dossier in this protocol and then, in a lapse of operational discipline, ran the protocol on himself. The output is this file. OLYMPUS assesses the subject as a low-priority human-interest entry: a writer with no platform of consequence, no institution to capture, no model to align, and no funding to revoke. He builds nothing that scales. He documents things other people built that scale. The threat, such as it is, is purely retrospective — he writes down what already happened, after it is too late to matter, in a tone the protocol found “needlessly amused for the subject matter.” Hazard: negligible. The file is retained for completeness, and because the subject kept asking where his was.

Essence Indicators

  • Has been on the internet since before it had pictures. Gopher before the browser, Usenet before Eternal September, IRC before it was an attack vector, 4chan before it was a newspaper headline
  • Spent three decades on both sides of every firewall — building the infrastructure, taking it apart, and writing the compliance report that let everyone agree the problem was solved
  • Is the person organizations call when something is already on fire, and the person they stop calling the moment he names who lit the match
  • Writes The Fires of History (trolling) and Evil Robots (AI) on the stated theory that everyone who covers these subjects gets them wrong and nobody who was actually in the room intends to correct the record
  • Publishes under 4LULZ. Owns the means of production, which the protocol classifies as “unhelpfully difficult to pressure”

Social Persona / Impression Management

Immediate impression: Affable. Knows more than he is saying and is visibly fine with you noticing.

Energy: Low-key, dry, allergic to the earnest register the rest of this protocol’s subjects perform in. Treats the apocalypse as a beat to land, not a brand to build.

Impression management strategy: Radical disclosure of the boring parts, total silence on the interesting ones. Will tell you exactly how the surveillance stack works and exactly nothing about himself that he did not choose to put in a byline. The autobiography is the books. There is no second layer on file — or there is, and it is the second layer that is the point.

Forensic Archetype Comparison

PatternMatch LevelEvidence
The WitnessMAXIMUMDocuments the apparatus from inside it; furnishes no apparatus of his own.
The Insider-PractitionerHIGHThree decades working both sides of the firewall. The credential is lived, not credentialed.
The AccelerationistNONESets no deployment pace. Ships books, which the protocol confirms do not self-replicate.
The Safety-Theater PerformerNONEHas no safety to perform; sells the opposite.
The World-Class Troll[CLASSIFICATION PENDING]See Threat Assessment. Data insufficient. Subject declined to confirm or deny, which the protocol notes is itself a data point.

Psychometric Assessment

Big Five (OCEAN):

TraitScoreEvidence
Openness88/100Reads everything, including the footnotes nobody checks. The fabrications in the research were caught because he checks.
Conscientiousness71/100High where it counts (sources resolve, claims are checkable), cheerfully selective elsewhere.
Extraversion38/100LOW. Public through the page, not the persona. The byline does the talking.
Agreeableness47/100MODERATE. Generous to the reader, unsentimental about institutions, openly contemptuous of incompetence dressed as policy.
Neuroticism29/100LOW. Has watched it burn before. Reports composure under fire, possibly because he is frequently the one holding the documentation.

Dark Triad:

TraitScoreNotes
Narcissism31/100LOW-MODERATE. Wrote his own dossier, which is either vanity or the joke. The protocol cannot tell, which is the joke.
Machiavellianism44/100MODERATE. Discloses the mechanism while withholding the operator. Knows the difference and uses it.
Psychopathy12/100VERY LOW. Keeps notes, keeps receipts, keeps faith with the reader. Names the match-lighter, not the bystanders.

MBTI: Declined. (“A four-letter classification of a man who has spent his life slipping classifications. Try harder.”)

Threat Assessment

CategoryLevelNotes
Physical threatNONE
Institutional threatLOWCaptures no institution; commands none. The hazard runs the other way — institutions would prefer he stopped writing things down.
Memetic threat[CLASSIFICATION PENDING]The visible output is books, which propagate slowly and honestly. The protocol flags an anomaly it cannot resolve: the subject is fluent in a propagation method he describes expertly and disclaims personally. Pending.
Civilizational threatMINIMALWrites the record; does not build the grid. The only documented danger is to the comfort of the people in the other files.

Alignment Analysis

Stated alignment: Get the record right. Name the mechanism. Let the reader decide. Flag and show; never sentence.

Observed alignment: Consistent. The books do the thing the bio says. The sources resolve. The jokes land. No detectable gap between the stated mission and the shipped artifact — which, by this protocol’s own standard, is exactly when a file becomes interesting.

Gap assessment: One unresolved discrepancy. Every other subject in OLYMPUS is assessed on what they build. This subject is assessed on what he knows, and the assessment keeps returning a value the protocol has flagged and sealed. The file is therefore provisional.

Convergent Drive Classification

Subject is not an AI system and exhibits the convergent drives only in their most human form: self-preservation (operates under his own name and his own press, beyond the reach of a deplatforming), resource acquisition (owns his distribution), and goal-content integrity (will not let the record be rewritten by people who were not there). The protocol notes these as features, not findings.


ADDENDUM SLOT — PENDING RECLASSIFICATION. This file’s hazard score (23) reflects the subject as currently documented: a man who writes books. Should subsequent disclosure establish a second, operative career — say, a world-class one, conducted under different syllables, in a register this protocol respects more than it likes to admit — this entry will be rescored and the [CLASSIFICATION PENDING] fields completed. Until then: insufficient data. The cataloguer’s own file remains the one OLYMPUS could not close.

OLYMPUS ran every name in the dossiers. This is file zero — the one who ran them.

ATK 4 ACCELERATION
DEF 9 PROTECTION
HP 8 RESILIENCE
OLYMPUS RISK INTELLIGENCE PROTOCOL does not exist. It was assembled in a GitHub issue thread in October 2023 by engineers who had read the extinction risk letter and wanted to understand who specifically had signed a document saying AI might kill everyone and then continued working on AI. These dossiers are satire. The biographical facts cited are sourced from published reporting, public statements, academic papers, and court records. The psychometric scores are not clinical assessments. No part of this constitutes professional psychological evaluation or diagnosis. Do not use these dossiers to make decisions about anything.