Private Intelligence Terminal

Questions

1. What is this?

MERIDIAN is a private intelligence terminal. It monitors open sources around the clock, resolves what it reads to canonical entities, and fuses the result into one daily briefing across six domains: geopolitics, economics, society, technology, health and regulation.

Each briefing commits to probabilistic forecasts, and every forecast that resolves is scored against what actually happened, on a resolution criterion set before the fact.

Things that are about to become important leave traces in open data before they become important. The gap is never data. It is fusion.

2. Where does the data come from?

From open sources, through their public interfaces: public APIs, RSS feeds, government and institutional publications, public data portals and public channels. The entities MERIDIAN tracks are countries, organisations and companies; it keeps no records about private individuals. By category:

• Instrument-measured — flight and aircraft tracking, satellite thermal and night-light anomalies, maritime chokepoint transits, GPS-jamming and internet-routing anomalies, connectivity measurement, wastewater surveillance, trade flows and market data.
• Institutional — wire services, government statements and travel advisories, health institutions, humanitarian reporting, sanctions and court records, corporate-ownership registries, academic output, exploited-vulnerability catalogs.
• Reflexive — large-scale media analysis, prediction markets, open-source monitoring channels, public attention measures.

3. Is state media used?

Yes, deliberately — and always labelled. One rule keeps that from becoming contamination:

• Attribution is mandatory. The alignment is named in the sentence that cites the outlet; a state-aligned source is never presented as neutral fact.

Coordinated framing across same-bloc outlets is detected and strengthens the manipulation assessment. It never raises a tier.

4. What does a tier mean?

A tier says how much independent evidence stands behind a signal, and of what kind:

• SIGNAL DETECTED — too little convergence to elevate. Logged, not written up.
• LOW — minimal convergence, weak corroboration. Kept for tracking, not actioned.
• MEDIUM — persistent signal but a narrow source base, or auto-downgraded after 48h without fresh Class A evidence.
• HIGH — strong convergence across source types with official or measured evidence behind it, but not yet the two measured source types CRITICAL needs. Chatter-only clusters cap at LOW.
• CRITICAL — requires physical, instrument-measured evidence. Narrative-only convergence cannot reach it.

5. What is a Brier score, and is ours good?

The Brier score measures forecast accuracy: 0.000 is perfect, lower is better. Guessing 50% on everything scores 0.250, so that line is the floor to beat, not a target. The scale is steep below it: a forecaster who is right 80% of the time, and says so, averages about 0.160; right 90% of the time, about 0.090. It rewards being confident and right, and punishes being confident and wrong.

The current score is read live from the public track record rather than printed here, because a number frozen at build time starts lying the next time a forecast resolves.

6. What does the access key give me, and what does it cost?

Nothing, during the beta. No card, no trial clock, no plan that starts charging when you stop paying attention.

7. What is NOT claimed?

• Not financial advice. No directive recommendations — nothing to buy, sell, hold or size. Directive financial language is rejected in code. Market data is an input, never an output.
• No real-time alerting. A daily rhythm with conditional intraday updates. No latency guarantee, no paging, no service level.
• Not equal depth in all six domains. Geopolitics-first, peripheral vision elsewhere; the society domain has no institutional-grade source, which caps what can be concluded there.
• Not a news wire, and not the primary record. It is an analysis layer over sources you can read yourself.
• Forecasting deals in probabilities, not predictions. A 70% call is meant to be wrong roughly three times in ten; resolved forecasts stay in the public ledger with their outcomes either way.
• No warranty. One operator's system, running as an experiment and provided as-is.

← Back to MERIDIAN