AI Systems and their Scalability – A Systematic Literature Review

Focusing on the technological innovation of Artificial Intelligence (AI) systems, this paper presents a systematic literature review to analyse the scalability of these systems. The rapid development and widespread adoption of AI technology in various domains requires an understanding of scalability as a critical factor for successful innovation and potential business success. However, the term 'scalability' is often used inconsistently. This study aims to clarify the definition of scalability of AI systems, identify specific components that contribute to scalability, and explore approaches to achieve scalability. We analyse the different components of scalability through a detailed review of the existing literature. Our findings reveal a multi-dimensional view, and we propose a refined definition of scalability that incorporates these components. We also discuss different approaches to improve scalability. Our study not only contributes to a better understanding among stakeholders, but also provides a structured framework for future research and practical applications in the effective scaling of AI systems.