Lenders can now use 'predictive insolvency' to identify that a borrower will go bankrupt before they even grant the loan.
April 1, 2026
Original Paper
Debt Architectures: Cognitive Vulnerability and Algorithmic Bias in Digital Predatory Lending
SSRN · 6275020
The Takeaway
Algorithmic scoring has reached a point where lenders can detect a person's future financial collapse with high accuracy. Rather than denying credit, some digital lenders use this insight to design 'debt architectures' that exploit the borrower's final window of liquidity through hyper-personalized interfaces and 'doom spending' triggers.
From the abstract
This paper analyzes the rise of digital predatory lending as the result of decision architectures and algorithmic systems designed to exploit users’ cognitive vulnerabilities. Over‑indebtedness emerges from the interaction between endogenous behavioral biases and exogenous technological factors, including interface design, hyper‑personalization, and opaque scoring models. The intensive use of big data enables providers to anticipate patterns of financial deterioration and, despite this knowledge