A Comprehensive Use Case of Decision-Theoretic AI: Fraud Review Escalation under Capacity Constraints
Year: 2026
Type: article
Abstract: This paper develops a comprehensive, self-contained use case that operationalises the decision-theoretic framework of Noguer i Alonso (2026) as reviewed and summarised in the preceding thread. Building on the anatomical review of output contracts, acquisition contracts, capacity gates, and the discrete/continuous routing-geometry dichotomy, and on the subsequent extraction of key lessons and the illustrative fraud-review example, the paper specifies a complete bank fraud-escalation system. It defines the action set and reward matrix, identifies the minimal output contract, constructs the net acquisition advantage, implements the capacity-constrained optimal gate, and contrasts boundary-focused versus strongest-signal routing under discrete versus continuous exposure decisions. A numerical illustration demonstrates the calculations. The use case translates the theoretical results into implementable design principles while remaining faithful to the scope and non-claims of the original framework. All references follow Harvard style. KSCCN article, 11 October 2026. Companion to the KSCCN review Decision Theory and Artificial Intelligence: An Anatomical Review of Output Contracts, Acquisition Geometries, and Epistemic Verticals (https://doi.org/10.5281/zenodo.23298429). Dr. Syed Muntasir Mamun, ORCID 0000-0001-6845-2853. The views expressed are those of the author.
Author: Mamun, Syed Muntasir
Institution: Ministry of Foreign Affairs, Dhaka
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W7221434853