BillDrift

That's exactly the distinction I'm trying to be strict about. A deterministic mismatch by itself isn't enough for me to call something valuable. The important part is whether the underlying evidence is clear enough that an operator can inspect it, understand why it was flagged, and decide whether it actually deserves investigation or action. We've already been able to surface reproducible discrepancies in validation, but I'm deliberately not turning those into claims like “money recovered” or “operators disputed X” before we've seen that happen in real customer/operator audits. So the next validation step is less “can the system find anomalies?” and more “which of these findings survive a real operator's scrutiny and are worth acting on?” I'd rather underclaim that part than make every mismatch look commercially meaningful.