The Methodology
01 — Epistemic Independence
The fatal flaw of contemporary conversational AI in financial analysis is premature consensus. When multiple personas or LLM calls share a common conversational history, they rapidly converge toward polite, agreeable compromises.
Bourse Chamber solves this structurally through blind independent readings in Stage 01. All nine seats receive the exact same Evidentiary Pack simultaneously, but formulate their initial analysis without visibility into the arguments of their peers.
02 — Structural Cross-Examination
Agreement is cheap; constructive friction is rare. Following independent readings, the Moderator detects the acute philosophical fault line (e.g., Benjamin Graham's net liquidation floor requirement versus Cathie Wood's exponential S-curve TAM).
03 — The Taleb Barbell Sizing Doctrine
Most retail market catastrophes are not errors of asset selection, but errors of sizing. In Bourse Chamber, the position sizing band is not an average of nine votes.
04 — F13 Answer Budget (13 per Day)
Cognitive friction requires scarcity. Free tier participants receive 13 expert answers per day, resetting at 00:00 UTC. One full bench deliberation consumes 9 credits; the remaining 4 credits empower the user to directly question or cross-examine the specific seats they disagree with most.
Voting and verdict closure are strictly free (0 credits), ensuring every commenced deliberation is permanently closed and archived to the ledger.