An escalation of superlatives and a lack of risk acknowledgment in a CEO's speech predicts bank failures with 100% accuracy.
April 25, 2026
Original Paper
Detecting Institutional Failure Through CEO Earnings Call Language: A Combined NLP-Financial Signal with Zero False Positives
SSRN · 6578962
The Takeaway
A linguistic signature called the Invincibility Triad identified seven major US bank collapses before they happened. The framework combined natural language processing with balance sheet data to achieve zero false positives. CEOs of failing institutions tend to use specific patterns of confident language to mask underlying rot. This indicator proved to be a reliable and tradeable signal for institutional collapse. Investors can now use AI to listen for the specific sound of a bank about to go under.
From the abstract
We present a combined natural language processing and balance sheet framework that detects institutional failure risk by analyzing CEO language patterns on quarterly earnings calls. Using a corpus of 173,000+ transcripts spanning 4,800 tickers from 2006-2025, we identify a universal linguistic signature-the Invincibility Triad-that precedes major U.S. bank failures: (1) systematic absence of constraint language acknowledging domain-specific risks, (2) elevated inevitability language expressing f