Loading latest threat data...
Awaiting First Sweep
The automated daily analysis runs at 11:00 AM EDT every day. Data will appear here after the first sweep completes.
Next sweep:
State of the World Score
Awaiting Data
Data will be displayed after the next automated sweep at 11:00 AM EDT.
Chart data will appear after at least one sweep is completed.
Daily Global Analysis Log
Analysis entries will appear after sweeps are completed.
How Scores Are Computed
The Global Threat Matrix uses the Gemini AI API via Netlify AI Gateway to perform a daily automated sweep of global geopolitical communications at 11:00 AM EDT. For each entity the system searches, strictly within the preceding 24 hours, for qualifying primary-source material: official statements, press releases and readouts from heads of state, foreign and defense ministries, and cabinet offices; daily press briefings and spokesperson remarks (White House, State Department, Pentagon, and their global equivalents); major diplomatic speeches and remarks at international forums; congressional or parliamentary testimony by executive-branch officials; verified official social media posts from primary government accounts; and state news wires reproducing an official statement verbatim.
If an entity returns no qualifying item on the first pass, the system performs a second retrieval pass with a broader prompt that still enforces the 24-hour window but accepts additional same-day signals — quoted wire-service reporting that cites the original government source, statements from other senior executive-branch officials speaking in an official capacity, and official readouts of same-day calls or meetings. An entity is only marked N/A once both passes decline.
The AI then acts as an expert geopolitical psychologist. It reads the retrieved statement and analyzes its emotional and political tone. Based on this psychological posture, it assigns a Threat Level from 0 to 100, where 0 represents complete cooperation and peace, and 100 indicates extreme hostility or imminent conflict. The Global Score is the mathematical average of these individual entity threat levels, excluding entities with no qualifying statement on that given day.
Automated Daily Sweep
Sweeps are fully automated. A Netlify Scheduled Function runs every day at 11:00 AM EDT. The results are stored persistently and accumulate day by day, forming a historical database. Visitors can download the complete dataset in JSON or CSV format from the Trend & History tab.
Limits of Analyzing Official Statements
- Performative Nature: Official statements are inherently performative. They are crafted for domestic political consumption, international posturing, or strategic ambiguity rather than absolute transparency.
- Deception and Masking: Entities may issue peaceful statements to mask impending aggressive actions, or use highly aggressive rhetoric merely as a negotiating tactic without intent to act.
- Cultural Nuance: AI analysis of translated or culturally specific rhetoric may occasionally miss subtle diplomatic subtext or historical context embedded in the phrasing.
- Availability Bias: The model relies on publicly available web data. Classified intents, back-channel diplomacy, and closed-door realities are entirely invisible to this metric.