A Secondary Study on the Effectiveness of Hybrid Human–AI Models in Enhancing Credit Risk Assessment in the Banking Sector with Special Reference to State Bank of India

Abstract

Credit risk assessment is a fundamental function in the banking sector, directly influencing financial stability and lending decisions. Traditional credit evaluation methods, primarily based on financial ratios, credit history, and human judgement, often face limitations such as subjectivity, slower processing, and inability to analyse large and complex datasets. With the rapid advancement of artificial intelligence (AI) and machine learning (ML), banks are increasingly adopting data-driven approaches to improve the accuracy and efficiency of credit risk prediction.

This secondary research paper aims to examine the effectiveness of hybrid human–AI models in enhancing credit risk assessment, particularly in the context of large public sector banks such as the State Bank of India (SBI). The study is based on an extensive review of existing literature, research journals, industry reports, and financial publications. It explores how AI-based models, including logistic regression, decision trees, and neural networks, outperform traditional methods in identifying patterns and predicting loan defaults.

However, the study also highlights the limitations of fully automated systems, such as lack of transparency, algorithmic bias, and reduced accountability. To address these challenges, the paper emphasises the importance of integrating human expertise with AI systems. A hybrid model enables banks to combine machine efficiency with human judgement, leading to more accurate, fair, and reliable credit decisions.

The findings suggest that hybrid Human–AI models can significantly improve credit risk management, reduce non-performing assets (NPAs), and enhance overall decision-making efficiency in the banking sector.