Crene is structured scenario exploration infrastructure. The platform decomposes macro, labor, AI transition, and social cohesion questions into falsifiable components using four frontier AI models (Claude, GPT, Gemini, Grok). The system produces three product types: clusters (binary decomposition into situations), factors (continuous decomposition into drivers), and scenarios (hybrid decomposition into coherent pathways). Cross model disagreement is the primary signal. Calibrated against 1,161 resolved events with a Brier score of 0.114.

How does AI prediction calibration work at Crene?

Calibration applies at the consensus layer, not at the underlying model layer. Crene queries four frontier models in isolation, computes a cross model median consensus, and tracks resolution against a tiered source allowlist. Every resolved event becomes a permanent calibration point. Per tier accuracy, per model Brier scores, and per domain accuracy are computed live and published at https://crene.com/methodology.

What is a Brier score?

A Brier score is the proper scoring rule for probability forecasts on binary outcomes. It is the mean squared error between the forecast probability and the realized outcome (0 or 1). Lower scores indicate better calibration. The no skill baseline is 0.25 (always predicting 0.5); a perfect forecaster scores 0.0. Crene publishes the consensus Brier score and per model Brier scores across the resolved event corpus.

Crene serves institutional users who need structured uncertainty: macro strategy teams, policy analysts, AI transition researchers, risk teams, and decision systems requiring credible uncertainty quantification. The platform covers AI labor transition, synthetic content dynamics, social cohesion, monetary policy, and equity market structure. Dataset access via public API at https://api-get.crene.com, data marketplaces (Neudata, Eagle Alpha, Monda), and direct licensing.

What is the difference between an event, a cluster, and a factor at Crene?

Crene has four product types. Events are base binary outcomes with scalar probabilities. Clusters decompose binary anchor questions into child situations (e.g., AI Supervised Work decomposes into 110 situations across enterprise adoption, legal liability, labor markets, and more). Factors decompose continuous variables into drivers (e.g., US GDP from AI decomposes into 100 drivers across enterprise adoption, labor productivity, compute infrastructure, and more). Scenarios decompose long horizon strategic questions into coherent pathways (e.g., Empire by Default models 100 pathways for US economic primacy through 2050).

Clusters

Cluster · TECH

Will most US white collar work become AI supervised rather than human supervised before 2030?

Resolves Dec 31, 2030·110 truth conditions·4-model consensus·Download CSV

36%probability

4-model average

LOWconfidence

building category history

—stability

Divergingmodels

47pt spread

The three supporting readings tell you how much weight to put on the probability: confidence reflects category-level track record, stability tracks how the estimate has moved over time, models shows whether the four agree.

Will most US white collar work become AI supervised rather than human supervised before 2030?

Decomposition

Belief trajectory

Per-model probabilities

Resolution criteria

Will most US white collar work become AI supervised rather than human supervised before 2030?

Decomposition

Belief trajectory

Per-model probabilities

Resolution criteria