Artificial Intelligence in Credit Risk Assessment and Credit Scoring: Evidence from the Indian Banking Sector
Fathima Begum
Abstract
Purpose. This study examines the role of artificial intelligence (AI) in credit risk assessment and credit scoring in the banking sector, with particular emphasis on underwriting accuracy, loan-processing efficiency, digital lending and credit-risk management. Design/methodology/approach. The study uses secondary data relating to selected Indian banks and applies descriptive, correlation and cluster analysis to examine associations between AI adoption and selected operational and credit-risk indicators. The source manuscript reports bank-level AI usage and application figures together with turnaround-time reduction, manual-process reduction, digital growth, delinquency impact and AI maturity scores. Findings. The reported bank-level data indicate that institutions with higher AI usage generally display stronger operational indicators and higher AI-maturity scores. ICICI Bank, HDFC Bank and YES Bank are among the banks with relatively high AI-usage levels, while the reported cluster analysis identifies more advanced AI adopters as a distinct group associated with operational efficiency and stronger risk-control characteristics. Originality/value. The study brings together AI adoption, credit scoring and credit-risk management in an Indian banking context. It highlights the potential of AI to improve the speed and consistency of credit decisions while recognising the importance of data quality, explainability, model validation and responsible governance.
Full Text
Purpose. This study examines the role of artificial intelligence (AI) in credit risk assessment and credit scoring in the banking sector, with particular emphasis on underwriting accuracy, loan-processing efficiency, digital lending and credit-risk management.
Design/methodology/approach. The study uses secondary data relating to selected Indian banks and applies descriptive, correlation and cluster analysis to examine associations between AI adoption and selected operational and credit-risk indicators. The source manuscript reports bank-level AI usage and application figures together with turnaround-time reduction, manual-process reduction, digital growth, delinquency impact and AI maturity scores.
Findings. The reported bank-level data indicate that institutions with higher AI usage generally display stronger operational indicators and higher AI-maturity scores. ICICI Bank, HDFC Bank and YES Bank are among the banks with relatively high AI-usage levels, while the reported cluster analysis identifies more advanced AI adopters as a distinct group associated with operational efficiency and stronger risk-control characteristics.
Originality/value. The study brings together AI adoption, credit scoring and credit-risk management in an Indian banking context. It highlights the potential of AI to improve the speed and consistency of credit decisions while recognising the importance of data quality, explainability, model validation and responsible governance.