Scenario literacy · not a prediction engine
IARPG uses systems thinking to make fictional consequences easier to inspect: what changed, why it changed, what evidence supports it, which assumptions matter, who is affected, and how a different decision could produce another branch.
This is a content method for missions, alliances, logistics, infrastructure, information environments, and institutional change. The site is not building a planetary simulator, nuclear-effects calculator, public-health model, digital twin, or live geopolitical forecast.
Direct answer
It means using explicit states, relationships, assumptions, evidence, and reversible events to explain fictional consequences. A model is selected because it fits a question—not because it looks impressive. Every public model needs a purpose, truth boundary, accessible representation, uncertainty statement, validation record, and a clear list of interpretations it does not support.
Choose the lens by the question
No single simulation form explains everything. Each family makes some mechanisms visible and compresses others. The publication therefore records what a model is for, what it hides, and what decisions it must not support.
What area or set of systems is affected under stated assumptions?
How might state move between categories over time?
How can local actor rules and relationships produce system-level patterns?
How did the state change, and what differs across alternative histories?
Which findings persist across different assumptions or model forms?
Model card before model output
A polished map or animation is not validation. Before any scenario result appears, the publication should expose the question, data, assumptions, state rules, uncertainty, verification, validation, rights implications, and correction history.
That structure follows the same principle already used by IARPG campaign turns: the current state is not enough. Readers need the prior state, reason, evidence, dissent, consequences, and reversal conditions.
Where the method improves IARPG
The reports are used to deepen existing pages and standards—not to launch a separate planetary game. Each application stays fictional, evidence-led, reversible where possible, and bounded against real-world profiling or targeting.
Explain how evidence quality, access, civilian impact, and institutional response change after a fictional operation.
Trace domain-specific cooperation, burden, dependency, dissent, correction, and reversible realignment.
Show dependencies, service loss, restoration, and unequal effects without exposing real targets.
Compare rumor, correction, confidence, and trust pathways without profiling real people.
Describe routes, bottlenecks, jurisdiction, and recovery as fictional system states.
Attribution correction
The supplied reports repeatedly expand IARPA as “Integrated Artificial Reality Planetary Atlas.” That is a fictional phrase, not the agency’s name.
The official name is Intelligence Advanced Research Projects Activity. Its public mission is high-risk/high-payoff research for difficult Intelligence Community challenges, and the agency states that it is not operationally focused. IARPG may cite public program descriptions as research context, but it must not imply sponsorship, endorsement, partnership, or an official planetary-simulation program.
Attribution and model-output boundary
Views, impressions, shares, forwarding, outages, and media pickup can describe exposure or disruption. They do not establish belief change, behavioral change, policy impact, sponsor intent, or strategic success. Model outputs must remain conditional and separate from observed evidence.
Preserve the event, source, date, scope, and uncertainty before modeling consequences.
Use verified distribution evidence and state its coverage limits.
Require experiments, credible quasi-experiments, or another explicit causal design before claiming effect.
Keep rules, parameters, sensitivity, alternatives, affected communities, and correction routes visible.
Eight reports, bounded for reuse
Each source is stored as an individual Markdown file. Each canonical report records source lineage, attribution corrections, reusable methods, rejected claims, safety restrictions, current-verification needs, and direct links back into active UAI memory.
Hard boundary
The repository keeps the useful architecture and rejects the unsafe interpretation.
Scenario is not prediction.
Visualization is not validation.
Precision must not exceed evidence.
Generated narrative cannot own authoritative state.
Real agency names and programs require primary-source verification.
Every model needs an accessible non-visual representation.
No model may create identity-based dangerousness, recruitability, targetability, or moral scores.
Additional boundary: the preserved source on nuclear-effects simulation remains archival evidence. Current IARPG content does not reproduce weapon-effect equations, casualty calculators, fallout models, target optimization, or deployment guidance.
From shock to recovery
Use the World Systems sequence to connect an initial condition with authority, domestic response, regional coordination, material dependencies, evidence quality, unequal impacts, correction, and recovery. The model remains conditional and must show disagreement, sensitivity, affected communities, and reversal conditions.
Conditional cascades, not forecasts
Each country proposal records formal response structures, selected affected-community evidence, unequal-consequence questions, recovery uncertainty, and a bounded fictional scenario translation. Audits and volatile rechecks keep source observation, model inference, reviewer evidence, editorial adjudication, and fictional translation distinct; no source recheck is relabeled as human judgment.
Scope, language, confidence, and review remain attached to the claim.
Rationale and a falsifiable follow-up stay visible.
Output is not relabeled as direct observation or prediction.
Invented actors and locations replace current communities and institutions.
Connected tools and standards
