Artificial Intelligence, Innovation and Growth in MSMEs:

A Structural Model usin PLS-SEM

Abstract

The purpose of this research was to assess the relationship among business growth, artificial intelligence (AI) adoption, and innovation in micro, small and medium-sized enterprises (MSMEs) in the commerce and services sector of Southern Sonora. A quantitative, causal, non-experimental, and cross-sectional study was conducted, temporally delimited between January and March 2026. The data collection instrument was structured in four sections operationalizing the variables through items adapted from existing literature and validated using a Likert-type scale. The sample comprised 370 valid questionnaires administered directly to business owners and managers through non-probabilistic convenience sampling. Data analysis was performed using structural equation modeling via partial least squares (PLS-SEM). The structural model results indicated that all three hypotheses were statistically significant (p < 0.001). The strongest relationship was identified between artificial intelligence and innovation (β = 0.574), confirming that technology functioned as a catalyst for the innovation process. Furthermore, innovation exerted a direct impact on business growth (β = 0.341), which exceeded the direct effect of artificial intelligence (β = 0.263). In conclusion, the model demonstrated predictive relevance and moderate explanatory power (R² = 0.289), providing evidence that innovation capacity functioned as the critical pathway through which technological adoption translated into measurable competitiveness improvements.