CounterBrain is the reference implementation of the cognitive contradictor: an adversarial AI system, conceived by Francesco Saverio Canepa, designed to challenge board and committee decisions rather than confirm them. It is the operational testbed of the two companion papers Cognitive Sovereignty and Calibrated Dissent (OSF, DOI 10.17605/OSF.IO/5QC8T).
Three elements characterise it. First, the Observatory: a public, append-only ledger in which every forecast is sealed before any outcome is observable, with a double anchor — SHA-256 and OpenTimestamps on the Bitcoin blockchain — so anyone can independently verify that a forecast existed before a given date. Second, the Δ-CSI: a computed index of the intensity of the challenge — never of the correctness of the decision. Third, the public pilot: ten forecasts on European M&A deals, sealed in June 2026, with outcomes maturing over 2027–2029 windows.
The pilot illustrates the protocol; it does not prove its efficacy: N = 10 is explicitly non-evidence, and confirmatory claims belong to a powered Phase 2 whose analysis plan is pre-registered. Details at counterbrain.com and in the Research section.
