← All Projects
Meridian · Open-Source Threat Surveillance

Snow

The information needed to anticipate most global threats is already public. Snow finds it before the crisis does.

Active · Beta · Live Ingestion · Infectious Disease · CBRN · Humanitarian
Infectious Disease · High Signal
Unusual pneumonia cluster reported across 3 provinces — 14 sources, 6 languages
Detected 18h before WHO advisory · Southeast Asia
Humanitarian · Medium Signal
Aid convoy access denied for 4th consecutive week — escalating coverage pattern
27 articles across 11 outlets · Horn of Africa
CBRN · Monitoring
Industrial facility evacuation reports — chemical odor, authorities investigating
Early watch · Eastern Europe · Unconfirmed
Overview

John Snow mapped the 1854 Broad Street cholera outbreak without germ theory, without a microscope, and without classified intelligence. He read what was publicly visible and drew the right conclusion in time to act.

Get Early Access

The signal was always
there. Snow finds it.

Snow is an AI-driven open-source intelligence tool that monitors global news and media feeds to detect and track non-traditional combat threats — infectious disease outbreaks, CBRN incidents, humanitarian crises, and civil unrest — before they escalate beyond the point where institutional response can be preventive rather than reactive.

The architecture is built on a principle proven in 1854: the information required to anticipate most global threats is already publicly available. It appears in local news coverage, regional health reports, social media posts from affected communities, NGO field updates, and government communications. The challenge is not access — it is signal extraction at speed and scale.

Snow reads across languages, sources, and geographies simultaneously. It identifies anomalous patterns — clustering coverage, escalating source counts, geographic spread, terminology shifts — and surfaces them as structured threat signals before the volume of coverage triggers traditional institutional alarm mechanisms.

Threat Categories

Non-traditional
combat threats.

Category 01
Infectious Disease & Pandemic Risk
Detection of unusual disease clustering, novel pathogen reports, antimicrobial resistance trends, and outbreak escalation signals before formal health authority notification. Built for the pre-alert window where early action matters most.
Unusual pneumonia clusters in regional media
Antimicrobial resistance escalation signals
Zoonotic spillover early indicators
Category 02
CBRN Incidents & Hazards
Monitoring for chemical, biological, radiological, and nuclear threat signals in open-source reporting — industrial accidents, evacuation reports, unexplained casualties, and regulatory anomalies that may precede or accompany CBRN events.
Industrial facility evacuation patterns
Unexplained mass casualty reporting
Regulatory and environmental anomalies
Category 03
Humanitarian Crises & Displacement
Early detection of escalating humanitarian situations — aid access denial, displacement acceleration, food insecurity signals, and civilian protection deterioration — before crises reach the scale that overwhelms institutional response capacity.
Aid access denial escalation patterns
Displacement corridor monitoring
Food insecurity leading indicators
Methodology

Open source.
Rigorous signal.

01 — Ingestion
Multi-source, Multi-language Feed Aggregation
Snow ingests news feeds, government communications, NGO reports, academic pre-prints, and social media signals across dozens of languages simultaneously. No classified sources. No proprietary data. Everything Snow knows is publicly available to anyone who reads fast enough.
02 — Detection
Anomalous Pattern Recognition
AI models identify anomalous patterns in coverage volume, geographic clustering, source diversity, temporal acceleration, and terminology shifts. The signal is not any single report — it is the pattern of reports across sources and time.
03 — Classification
Threat Typing and Severity Assessment
Detected signals are classified by threat category and severity level. Snow distinguishes between noise, monitoring-level signals, and actionable threat indicators — reducing false positive burden on analysts while ensuring genuine signals surface.
04 — Output
Structured Intelligence Products
Snow outputs structured threat signals with source attribution, geographic context, and confidence levels. Designed for integration with existing institutional workflows — not a replacement for human analyst judgment, but a force multiplier for the analyst's attention.
Named For

"I feel confident that restricting the use of the pump in Broad Street... will have a great effect in diminishing the calamity."

John Snow, letter to the Board of Guardians, 1854

In August 1854, cholera swept through the Soho district of London. The prevailing theory was miasma — that the disease spread through bad air. John Snow disagreed. Using nothing but door-to-door interviews, local records, and a map of Broad Street, he identified the Broad Street pump as the source and persuaded authorities to remove its handle.

He had no microscope. He had no germ theory. He had public information, rigorous pattern recognition, and the clarity to act on what the data showed rather than what the prevailing theory predicted.

Snow the product is named for that methodology. The information needed to anticipate most global threats is already public. It requires the right instruments to read it in time, and the institutional courage to act on what those instruments reveal.

Also from Meridian

The full portfolio.

Epistemic Infrastructure
Cronkite
Real-time epistemic scoring for any web content. Five-model AI ensemble. Crowdsourced domain trust database. The ambient immune system the information environment never had.
Country Risk Intelligence
Marshall
Living country risk scores across a structured global risk taxonomy. Updated continuously. Built for policy makers, NGOs, researchers, and institutional decision-makers.

Meridian is the independent research and applied-technology project of Mark Greenhalgh. Views, analysis, and materials published under the Meridian name are his own and do not represent the position of the U.S. Department of Defense, the Department of the Army, or any U.S. government agency or component. Meridian operates independently of his official duties.