AI Documentation

Written From Records. Not From Memory.

Eleven finished intelligence products — from a threat brief to an evacuation plan. Each is grounded in current intelligence, aligned with the live risk picture, and delivered as a governed, branded document ready for action.

Eleven Products. No Template to Hunt For.

Ask in plain language in the intelligence chat, or let the mission workflow produce them for you as the mission runs.

Intelligence Brief

Regional intelligence brief written for the Overwatch ops room — key facts, risk picture, recommended actions.

Risk Assessment

Structured risk evaluation anchored to the live country-risk baseline — queried first, never guessed.

Entity Profile

Entity profiling to intelligence community standards — threat actors, organisations, and locations of interest.

Travel Dossier

Pre-travel package with mission context and verified emergency contacts — if a contact can’t be verified, it says so instead of inventing one.

Intelligence Digest

Comprehensive digest with executive summary, key events, trends, and recommendations across your areas of interest.

Evacuation Plan

Full evacuation plan with facility-validated extraction points, built on your organisation’s own doctrine.

Pattern Detection

Systematic pattern detection and behavioural analysis to intelligence community standards.

Route Analysis

Route-level risk assessed on realistic journey conditions — not straight-line guesses.

Event Analysis

Senior-analyst-grade synthesis of a single event — what happened, what it means for your people and sites, and what to do about it.

7-Day Threat Forecast

Seven days ahead — where risk is building and where it is easing, drawn from current signals and historical patterns.

Trip Debrief

Post-mission record built from check-ins, alerts, and the full mission timeline — frozen once the mission closes.

01 · Evidence In

PROVENANCE

What Separates Threatwhere From a Chat Window

Anyone can bolt a model onto a search box. The hard part is a document that stands up when the decision is challenged — and that is an engineering problem, not a prompt.

Governed Intelligence Evidence

Written From Records, Not Recall

Threatwhere writes from governed intelligence evidence, not model memory. Every document stays anchored to records your team can review.

Retractions Propagate

Retract intelligence once and it stays out of every future deliverable — automatically.

Fail Closed, Never Fabricate

Safety-critical fields fail closed: if an emergency contact or an extraction point cannot be verified, Threatwhere refuses to present one — it will not fill the gap with something plausible.

Live Country-Risk Baseline

Risk language is anchored to the platform's own live scored country-risk baseline — queried at the moment of writing, never estimated, and never inherited from a static advisory that is months out of date.

LIVE GENERATION

Watch a Real Assessment Take Shape

This is genuine platform output — a risk assessment written from retrieved intelligence, scored and sourced. Leave the page if you need to; the finished document is saved and waiting when you return.

TWgenerating

Generation

  1. ·Retrieving intelligence records
  2. Anchoring regional risk baseline
  3. Profiling threat actors
  4. Correlating recent incidents
  5. Scoring likelihood and impact
  6. Building mitigation guidance
  7. Governing output for release

Risk Assessment

Terrorist attack on military convoy near Bamako, Mali

Retrieving intelligence records

Scoring likelihood and impact against the live risk baseline…

02 · Governance Through

Your Doctrine. Your Plan.

Your organisation's own doctrine shapes the plan — governance and decision authority, extraction policy, family guidance, aftercare, communications cascade, and business continuity — written into supported products alongside your preferences, from risk tolerance to check-in cadence.

OUTPUT GOVERNANCE

Governed on the Way Out

Every generated product is governed before release: deliverables protect sensitive collection, tooling and attribution details while preserving the intelligence the reader needs to act. What leaves the platform reads as institutional analysis.

Organisation Doctrine

Governance & Decision AuthorityExtraction PolicyFamily GuidanceAftercareCommunications CascadeBusiness Continuity

Preferences

Risk ToleranceCheck-in Cadence

Governed Before Release

  • Sensitive CollectionProtected
  • ToolingProtected
  • Attribution DetailsProtected

Reads as Institutional Analysis

03 · Decision Out

From Operating Picture to Finished Product

Click a threat marker on the map and hand it straight to the AI. Generate reports from any event, any incident, or the travel-risk workspace — the same way, everywhere. Geofences generate intelligence reports — and generated reports create geofences. One platform, two licence types: every product is available in the Overwatch ops room, and every Overwatch licence includes an Operator licence for the field.

Operating Picture

Threat MarkerAny EventAny IncidentTravel-Risk Workspace

Finished Product

The same way, everywhere

GeofencesGenerated Reports
Overwatch Ops RoomOperator Licence Included

Written Is Only Half the Job.

A document only counts when it reaches the person who acts on it — with the branding and the permissions that survive scrutiny.

Share with Permissions

Named colleagues at view, comment, or edit — with an in-app notification when it lands.

Email as Branded PDF

Email the branded PDF straight to stakeholders — with a deep link back into the platform.

Print / PDF

Print or export to PDF in your organisation’s own branding, with a report-type banner on every page.

Files Itself

Reports auto-save on completion — mission reports land in the mission’s team workspace; standalone research stays private to you.

Ask Out Loud. Read the Full Picture.

Speak the question. Hear the briefing while the full document appears on screen — ready to review, save, and share like any other intelligence product.

0

Intelligence product types

Doctrine-led

Evacuation plans built on your own SOPs

Fail-closed

Refuses to fabricate — by design

Governed

Output governance on every document

11 products · retrieval-grounded · governed output

Stop Drafting. Start Deciding.

Eleven intelligence products, written from retrieved intelligence, governed on the way out, and delivered as branded documents your stakeholders can act on.