Causal AI Demo
Cause & Effect Reasoning with the do-Operator
A structural causal model built as a directed acyclic graph (DAG). Nodes are variables; edges are direct causal effects.
Click any node to apply an intervention, do(X = x), watch the effect propagate downstream, then click again to remove the intervention and return to observational reasoning.
Green edges are positive causal effects, red edges are negative. Node colors encode role: purple = root (exogenous), green = mediator, red = outcome. The right panel shows the active interventions and the directed causal paths from any highlighted node to the revenue outcome.
Built with D3.js for the DAG layout and rendering. The model uses structural equations over the graph's topology, so intervening on a node severs its incoming edges and forces its value while downstream variables recompute.