Home

The Global Curiosity Engine (GCE) is a protocol for locating where long-horizon predictive certainty begins to
fail. It does not explain causes, offer forecasts, or recommend actions. Instead, it applies falsification-first
checks (window perturbation, threshold sensitivity, and cross-dataset or cross-revision comparison) to surface
“Predictive Stress Zones” where trend continuity breaks. Its only output is disciplined human inquiry: concise,
assumption-focused questions that mark the boundary between what can be predicted reliably and what cannot.

GCE is a structured, repeatable way to scan a dataset that people (scientists, policymakers, planners, modelers) are actively relying on for long-term decisions, and it flags where hidden flaws or breaks in the assumed smoothness/trend continuity are real enough to demand human attention—before those flaws propagate into bigger surprises or bad allocations.