This paper assesses the financial distress management in the public sector and explores the application of Artificial Intelligence (AI) techniques to the Portuguese municipalities (n = 308). The paper aims to analyse the financial structure of these municipalities by identifying latent factors that characterise their economic and financial performance. The research question addresses which accounting variables provide the most robust explanatory power for managing financial distress. A quantitative methodological approach is adopted, employing Exploratory Factor Analysis (EFA) on variables extracted from municipal financial statements, namely the balance sheet and the statement of income by nature. Data is sourced from the Portuguese Accounting Standardisation System for Public Administrations and covers the most recent available fiscal year. EFA enables the reduction of data dimensionality, revealing principal components that group correlated financial indicators into interpretable factors. These factors are expected to distinguish between financial robustness and exposure to financial risk. By demonstrating the potential of data-driven models for public financial management, this paper sets the groundwork for extending AI-based financial distress detection frameworks to other institutional contexts beyond local government, including corporate and non-profit sectors. The findings aim to contribute to enhanced financial governance, transparency, and early warning systems supported by AI technologies.