A Multidimensional Evaluation and Comparative Analysis of Artificial Intelligence Application Popularity: Evidence from the UK and US Markets

Abstract

With the rapid advancement of artificial intelligence (AI) technologies, the adoption and popularity of AI applications vary substantially across national markets. Systematically characterizing and comparing these popularity patterns has therefore become an important research issue. This study develops a multidimensional evaluation framework encompassing technological performance, functional and scenario adaptability, market ecosystem, sociocultural environment, and network effects. Using this framework, the entropy-weighted TOPSIS method is applied to conduct a comprehensive evaluation and ranking of twelve mainstream AI applications in the UK and US markets. In a research context characterized by a small sample size and multiple heterogeneous indicators, the study provides an exploratory assessment of the relative popularity of AI applications.

The results reveal two main findings. First, clear heterogeneity exists across application types: text-based AI applications generally achieve higher composite rankings, whereas image-generation applications display a more dispersed distribution of scores. Second, when benchmarked against user adoption shares in the UK and US markets as a proxy for real-world popularity, the composite evaluation results show a high degree of consistency with observed usage patterns. This consistency suggests that the proposed multidimensional evaluation framework is capable of capturing key features of AI application popularity in practice.

By adopting a composite evaluation perspective, this study avoids the limitations of causal inference under small-sample conditions and offers a quantitative reference for comparative studies of AI applications and market positioning strategies.