ARTIFICIAL INTELLIGENCE-DRIVEN FINANCIAL INNOVATION FOR INTELLIGENT AND SECURE FINTECH SERVICES

Artificial intelligence (AI) is increasingly being incorporated into financial systems to strengthen risk assessment, analytical decision-making, and intelligent FinTech services. Predictive modelling can help identify financial patterns associated with corporate distress and provide additional information for risk-oriented decision support.This study aimed to evaluate AI-based predictive models for corporate bankruptcy prediction and examine their potential relevance to intelligent financial risk management and FinTech decision support.The study used the publicly available Polish Companies Bankruptcy dataset obtained from the UCI Machine Learning Repository. The third-year forecasting dataset was analysed using Logistic Regression, Random Forest, and HistGradientBoosting. Financial variables were prepared using median imputation, and the data were divided using an 80:20 stratified train-test split. Model performance was assessed using accuracy, precision, recall, F1-score, ROC-AUC, and PR-AUC.The analysis identified noticeable differences between bankrupt and non-bankrupt companies across profitability, leverage, liquidity, equity, working capital, and cost-related indicators. The evaluated models showed different predictive capabilities, with the tree-based approaches demonstrating stronger performance across several evaluation measures than the linear benchmark. The results also demonstrate the importance of using multiple evaluation metrics when analysing imbalanced financial data.AI-based predictive modelling provides a useful analytical foundation for financial risk assessment and intelligent decision support. Its application in FinTech can support early identification of financially vulnerable organizations, although external validation, explainability, governance, and continuous monitoring are required before practical deployment.