Almost everything we’ve published so far sits in the United States. That is a limit of habit, not of method — the world model carries regions everywhere. So we pointed it south, at the state of Rio de Janeiro, and asked a synthetic population of ~17.9M a single hypothetical question: fund a state-wide transit expansion through a temporary levy, yes or no?
Why Rio, first
Rio is a useful stress test. It concentrates dense urban core, sprawling periphery, and interior municipalities inside one state boundary — the kind of internal variance that makes a single number nearly meaningless and an open-ended “why” essential.
| Dimension | What it stresses |
|---|---|
| Urban vs. interior | Very different transit realities |
| Income spread | Who a levy actually lands on |
| Trust in institutions | Whether “temporary” is believed at all |
The one thing that dominated
Support for the idea was high. Support for the mechanism was not.
“A expansão eu quero. O imposto ‘temporário’ eu já conheço.” — synthetic voter, Baixada Fluminense
(“I want the expansion. The ‘temporary’ tax, I already know how that goes.”) The gap between wanting the outcome and trusting the funding path was the single widest split in the run — wider than any urban/interior divide.
The point of going south
This is a first look, deliberately small. The reason to run it is calibration: the same method, a very different community, so we can start to see where the model’s priors are well-grounded and where they are thin. Watch the globe swing below the equator on the way in — the south is where we have the most still to verify.
The Rio electorate here is synthetic and the referendum is hypothetical. Figures describe a simulated population, not a poll of real residents.