Generative Artificial Intelligence in Finance

10Role of Generative AI for Fraud Detection and Prevention

Prasanna Kulkarni^*, Pankaj Pathak, Samaya Pillai and Vishal Tigga

Symbiosis Institute of Digital and Telecom Management, Symbiosis International (Deemed University), Pune, Maharashtra, India

Abstract

Today’s digital era manifest by the widespread use of digital transactions reflects in our heavy reliance on the Internet for activities such as mobile banking, social media, e-learning, entertainment, and digital commerce. With the convenience brought by these technologies, there is a growing concern about their potential misuse for fraudulent activities, outpacing the effectiveness of traditional fraud detection methods. While traditional methods often fail to recognize subtle anomalies in financial datasets, generative artificial intelligence (AI) can reveal them using variational autoencoders (VAEs) and generative adversarial networks (GANs). Its capacity to create synthetic data protects the privacy of the real data and also enables rigorous testing of algorithms to be used for detection. The chapter explores how generative AI can imitate genuine financial transactions. Such mock data are useful for training and refinement of fraud detection models.

The predictive capabilities of generative AI help in forecasting latent frauds. Historical data analysis supports generative models to infer patterns and identify possible weaknesses in existing financial systems. Generative AI thus proactively enables anticipating swindlers ...