Blinded comparison
Medium statements are compared against matched AI-generated zero-information controls, then scored without asking sitters to accept any theory in advance.
A beta research platform for anomalous communication claims.
Signal Beyond tests whether medium-generated statements contain target-specific information that exceeds matched AI-generated zero-information controls under blinded scoring. Survival is a hypothesis; anomalous communication is the empirical target.
The current Signal Beyond workflow has moved through live protocol refinements and focused production testing. Version 17 improves phone-friendly microphone guidance and protects sitter scoring progress while participants move through longer sessions.
Medium statements are compared against matched AI-generated zero-information controls, then scored without asking sitters to accept any theory in advance.
Recent updates support one, multiple, or unclear apparent communicators while preserving meaningful names, relationships, and attribution cues for scoring.
A public sitter registration flow is live so the project can collect cleaner sessions and expand beta testing without turning the claim into a belief test.
The protocol is designed to ask a narrow empirical question: do specific statements contain target-relevant information that survives controlled comparison? A positive result would still leave multiple explanations open, including survival, psi-mediated access, unknown cognitive or informational mechanisms, artifact, leakage, or no durable effect.