ARTIFICIAL INTELLIGENCE-BASED ADAPTIVE DATA ARCHITECTURE FOR SCALABLE LOW-CODE INFORMATION SYSTEMS

Abstract

The rapid development of artificial intelligence (AI), cloud computing, data-intensive applications, and low-code development platforms is changing the architectural requirements of modern information systems. Low-code technologies reduce the amount of manually written software code and enable a wider group of users to participate in application development; however, scalability, data quality, interoperability, governance, security, and maintainability remain significant architectural challenges. These challenges become more complex when artificial intelligence is incorporated into low-code environments because AI-based components depend strongly on the quality, availability, structure, and continuous evolution of data. This article proposes a conceptual model of an artificial intelligence-based adaptive data architecture for scalable low-code information systems. The proposed architecture combines AI-assisted schema management, automated data-quality assessment, metadata-driven adaptation, intelligent data integration, machine-learning monitoring, and federated governance. The research is based on a structured analysis of scientific literature, established software-engineering principles, data-quality frameworks, AI risk-management guidance, and contemporary research on low-code/no-code development. The analysis indicates that an adaptive architecture should treat data quality, scalability, interoperability, AI lifecycle management, and governance as interconnected architectural concerns rather than independent technical functions. The proposed model introduces an adaptive control layer that continuously evaluates workload characteristics, data-quality indicators, model performance, and system requirements and subsequently recommends or initiates controlled architectural adjustments. Such an approach can improve the ability of low-code information systems to evolve without requiring continuous redesign of their underlying data infrastructure. The article concludes that AI should not merely be embedded as an application feature in low-code systems; instead, it should become an architectural mechanism supporting data management, quality assurance, optimization, and controlled evolution.