Reflexive Demand Simulator

Frontier-token demand is driven by a few unbounded, long-horizon tasks whose spend feeds their own revenue. Set the loop gains, drop a shock on the timeline, and watch the boom or the bust play out.

Built on Giovanni Cattani, “Nobody is talking seriously about AI demand” (X, 1 Sep 2026). Quarterly model on a rolling horizon; inputs are refreshed daily by an autonomous pipeline. All numbers are scenario estimates, not forecasts.

Compare & share

Frontier token revenue · run-rate two years out

—

Where the tokens go run-rate, $B per year · actuals to the left of now, scenario to the right

Task horizon

80%-success horizon, hours of human work (log)

Capital-market conditions

index, 1.0 = neutral

Compute committed

$B of capex per quarter

Capacity utilisation

demand ÷ monetisable capacity

The loop spend → output → capital → bigger budgets

Quarter

Segment	Share	Loop gain	Multiplier	Growth q/q

Shocks click a quarter, pick a trigger

What underpins this scenario

What changed the pipeline’s last decisions

Sentiment gauges refreshed daily where a feed exists

How it works