The Five Beacons Model: A Prudential Architecture for AI Explainability and Legal Liability in Banking

The accelerated integration of Artificial Intelligence (AI) into core banking functions has generated a structural tension between algorithmic opacity and established prudential and liability frameworks. This paper introduces the Five Beacons Model (5B), a vertical, liability-oriented governance architecture designed to ensure decision-centric traceability across five institutional nodes: machine interface, engineering layer, human supervision, corporate governance, and client communication. By reframing explainability as an evidentiary infrastructure, the 5B model aims to mitigate legal exposure and strengthen systemic resilience within the Supervisory Review and Evaluation Process (SREP) and the Internal Capital Adequacy Assessment Process (ICAAP).

A conceptual and prudential addendum to the Five Beacons Model as developed in the monograph: Garcia-Maceiras, JM (2026). 'The Banking Risk of AI Explanation: The Five Beacons Model'. ADR Notebooks No. 1. Zyphrum Alchemists.