AI Bubble Burst Probability & Scenario Simulator

Aggregated Burst Probability

14.0%

0.0% vs Baseline

Horizon Hazard Rate

1.24%

Monthly failure density

Primary Dominant Trigger

Revenue Miss

35.0% likelihood

Cascade Amplification Factor

1.18x

Coupling multiplier

Interactive Probability Cascade Tree Drag nodes • Toggle link dependencies

Evaluation Timeline Horizon:

12 Months

MODEL VERDICT & SCENARIO CAPTURE ACTIVE STATE

Under the 12-Month Horizon, the implied probability of an AI bubble burst is 14.0%. The highest driver is Major AI Cloud Revenue Miss with an individual probability of 35.0% and cascade factor of 1.18x.

Understanding the local weighted probability model

Your browser does not support video.

Read the explanation

The local model begins with four assigned probabilities and weights. Revenue contributes point one four zero, hardware point zero six six, injunction point zero three six, and power point zero two eight. Their sum is point two seven, divided by weights totaling one. These assumed inputs are not calibrated forecasts or current market evidence. The graphic shows a weighted average, not a union probability or a simulation of real financial contagion. Three default coupling links are active. The code counts them and applies one plus three times point zero four five, giving one point one three five. The identities and direction of the active links do not change the numerical contribution; each checked link contributes equally. This diagram therefore distinguishes the drawn network from the actual count based calculation. It does not establish a learned dependency between the economic events. The horizon changes an exponential adjustment to the weighted score. Keeping the default inputs and three couplings, the formula gives about twenty one point nine seven percent at three months, twenty six point zero six percent at twelve, and twenty nine point eight one percent at thirty six. The moving point follows that exact code equation. Increasing this horizon is not new market evidence. The output is clipped between one and ninety five percent before the monthly hazard calculation. The stored initial label says fourteen percent, but computeModel replaces that with the formula result, about twenty six point zero six percent. At twelve months, one minus the twelfth root of survival gives about two point four eight five percent monthly hazard, under a constant hazard interpretation. The dominant trigger is chosen by highest raw input probability, not weighted contribution. Export saves these local assumptions. Offline D three initialization prevents the normal header bindings, so our selected slider checks do not prove a live complete graph or a validated forecast.