ARBITER Africa Risk Intelligence — Coherence-Ranked Regional Signals

Configure

Region · Query · Candidates

Query — chars

Best results frame around indicators of the desired outcome with orthogonal vocabulary across candidates.

Candidates 0 lines

Each line scored independently. Use ` — ` (em dash, space-padded) between headline and detail.

Coherence Ranking

Ordered by ARBITER score

idle

Ready. Choose a region preset, or paste your own corpus, then run coherence ranking.

Endpoint Log POST · public/compare

api.arbiter.traut.ai/public/compare · awaiting query

Signal Surface 15°N · 5°W

Mali · Burkina Faso · Niger · Chad · Mauritania

Active Region

Sahel. Mali, Burkina Faso, Niger, Chad, Mauritania. Multi-actor conflict environment with competing jihadist groups, ethnic militias, military regimes, and external power competition.

What This Proves

Coherence ranking turns a pile of signals into a priority order.
Same engine, same API — different domains, different corpora.
Air-gap capable for sovereign and government deployment.
Triage layer for analysts, not a replacement for them.

Boundary

ARBITER ranks geometric coherence. It does not predict outcomes, verify sources, or render operational decisions. The analyst remains accountable.

Configure

Situation · Actors · Frames

Situation — chars

Vocabulary-neutral description of the situation. Avoid policy language that pulls every actor toward one frame.

External Actors 0 selected

Each actor's doctrine prefix is fused with the situation. Same five orthogonal frames are scored against each lens.

Perspective Output

Per-actor coherence across five orthogonal frames

idle

Same situation. Different lenses. Each actor reads the situation through their own doctrine. Divergence detection flags when external powers disagree maximally on what the situation is.

Endpoint Log POST · public/compare · per-actor

awaiting actor selection

Perspective Field 26 actors · 5 frames

Geometric perspective inversion across external power lenses

The Capability

One situation. Run through ten different doctrine prefixes. Five orthogonal frames per run. The geometry shows which frame each actor finds most coherent — without telling ARBITER what the actors should care about.

Divergence Detection

When the gap between top-frame coherence across actors exceeds a threshold, the situation is contested. STABLE means actors converge on interpretation. CRITICAL means they read fundamentally different situations.

Use For

Briefings to clients with multi-stakeholder exposure.
Diplomatic risk modeling for sovereign clients.
Workshops — hand a laptop to a delegation, watch the framing flip.
Detecting when consensus narrative is breaking down.

Configure

Region · Question · Group set

Analyst Question — chars

Frame around indicators of the behavior being assessed — expansion, consolidation, recruitment, fragmentation, alliance shift.

Sub-State Actors 0 lines

One actor per line. Actor name — descriptor / region / current posture. Each is scored against the analyst question.

Actor Coherence

Ranked by alignment with the analyst question

idle

Sub-state actor ranking. Score militias, factions, ethnic blocs, and informal power structures against the same evidence question. Surface which actor best fits the trajectory being assessed.

Endpoint Log POST · public/compare · actors

awaiting actor set

Sub-State Surface below the state

Most firms map governments. DefSEC maps below them.

DefSEC's Specialty

Sub-state actors are the work. Militias, factions, ethnic blocs, informal power structures — the entities that actually shape conflict dynamics. Most consultancies look at governments. DefSEC looks below them.

What ARBITER Adds

Rank N actors against the same analyst question.
Surface which actor's profile best fits a trajectory.
Compare actor sets across regions on a normalized axis.
Re-run hourly as new evidence enters the corpus.

Boundary

The ranking reflects coherence with the question's vocabulary, not predicted future behavior. Field analysts validate, not the engine.

Configure

Absence frame · Routine baseline

Absence Query — chars

Describe the absence. What normally happens that has stopped. The query frames silence as a coherent semantic genus, not as keyword absence.

Routine Baseline 0 lines

Each line is a routine activity — described as if normal. ARBITER scores how coherent each is with the absence frame. Under silence query, normal scores low, anomalous absence scores high.

Absence Coherence

High score = activity most anomalous if absent

idle

The dog that didn't bark. Most intelligence is signal-present. Silence is signal-expected-but-missing. ARBITER measures coherence between an absence frame and routine activities — the most expected activity is the most anomalous if it disappears.

Endpoint Log POST · public/compare · silence

awaiting absence frame

Negative Space absence detection

What is missing is often louder than what is present.

Why This Matters

Every major strategic surprise involved structured absence. Russian buildup before Ukraine. October 7 preparation. Sahel coups before announcements. Triggers fail when monitoring signals-present and the operation moves to silence.

The Geometry

ARBITER does not match keywords. The query describes a world where normal things have stopped. The candidate that describes the most normal thing is most structurally incoherent with that world — but only if the missing-routine candidate is in the corpus.

The Test

No absence words in candidates.
Routine activity described as if normal.
High coherence under silence query = anomalous absence.
Low coherence = activity proceeding as expected.

Rank what deserves analyst attention, before reading the report.

Three live corpora. One coherence engine.