How it works

From a one-line idea to a full briefing in about a minute.

Behind the velvet curtain it's not magic, it's a structured AI pipeline that forces a single policy idea through the same lenses a serious policy team would use.

1

You describe the country and the policy

You write the country in question and the policies you'd like to enact, in plain language. No technical schema, no jargon. The simulator reads it the way a senior advisor would and asks itself the right follow-up questions.

2

The model thinks across domains

A reasoning model considers the proposal from multiple angles at once including: economics, social fabric, political feasibility, environment, geopolitics, civil liberties. It looks for both intended effects and the second-order consequences people miss.

3

Each claim is grounded and weighted

Every effect comes with an explicit probability and a magnitude (minor, moderate, major, transformative). Where relevant, the model anchors itself in real historical precedents — actual countries that tried something similar, and what actually happened.

4

A structured report comes back

You get an executive summary, a net outcome verdict, a risk assessment, a timeline of likely events across short and long horizons, a stakeholder breakdown, and the recommendation a non-partisan advisor would put in front of you.

5

You share, debate, refine

Every report has its own permanent link. Send it around. Argue with it. Then change one variable and run it again. That iteration is the point.

Want the technical version?

See methodology for the model, the prompt design, and the calibration approach.