ECONOMIC CONSEQUENCES OF PREMATURE OBSOLESCENCE IN LOW-CODE AND AI DEVELOPMENT

Abstract

Information systems in the digital economy are increasingly replaced or abandoned within 2-5 years of deployment-far shorter than their intended lifespan. This article investigates premature system obsolescence as an economic problem rooted in algorithmic and data structure decisions made at the design stage. Through a systematic analysis of 24 industry reports and academic sources, we demonstrate that incorrect algorithm selection - particularly in Low-Code/NoCode (LC/NC) platforms and AI-Driven Development (AIDD) environments - is a primary structural cause of performance degradation, technical debt accumulation, and costly system replacement. Our empirical findings show that systems built with O(n²) sorting algorithms degrade to unacceptable performance thresholds within 2.8 years on average, compared to 12.5 years for systems using optimal O(log n) structures. The economic cost differential reaches $4.8M per incident for worstcase algorithm choices vs. $180K for optimal implementations. We argue that LC/NC and AIDD paradigms, while accelerating development speed, systematically eliminate the developer's ability to make informed algorithmic decisions, thereby creating a structural vulnerability that manifests as premature obsolescence. Recommendations for algorithm-aware development frameworks are proposed.