The future is not a prediction.
It is a set of choices.
Build a miniature 2035 city. Change its priorities. Watch consequences ripple through people, institutions and streets. Every result is inspectable; none claims to foretell the world.
Enter the observatory ↓World simulator
Five issues. Three plausible futures each. One deliberately simple, inspectable model.
Renderer: initializing
Why this happened
Explicit input → rule → effect trace
Snapshots & exports
Save the current state locally and compare it to later decisions.
Five lenses for responsible imagination
Ideas do not live alone.
These paths connect the model to publicly accessible starting points. A link is a doorway, not evidence that the model is correct.
What this machine can—and cannot—tell you
ILLUSTRATIVE SIMULATION: Every city block, index and consequence is generated from authored rules and adjustable assumptions. The rules express contestable judgments, not estimated causal coefficients. The model has no live feeds, no trained predictive AI and no claim to empirical calibration.
ARGUMENT: Policies often trade one good against another; feedback, implementation and distribution matter. The future narratives are competing interpretations of the same toy system. Their purpose is to improve questions, not to prescribe policy.
PUBLIC SELF-DESCRIPTION: This is an independent creative technology experiment by Sahil Tayal. Explore the wider work at the central atlas ↗.