Evaluating the Role of Deep Learning and Human–AI Collaboration in Improving Anti-Money Laundering Compliance
Main Article Content
Abstract
The escalating intricacy of monetary dealings has brought on a strong requirement of topnotch Anti-Money Laundering (AML) systems which can even spot the suspicious activities in a very short time.A real-time detection has now become a necessity for financial transactions hence the technological development in Aml systems is the solution to minimize the risk.The current investigation is developing a decision-making configuration that will eventually involve humans andAIs in the department of finance thereby increasing trust and reliability. For this purpose, a mock dataset containing 100 observations was created for carrying out statistical evaluation, which would help to understand how AI model accuracy, false positive rate, human decision accuracy, collaboration scores, and finally, compliance improvement are all related to each other. The descriptive analysis and correlational study revealed very strong positive ties (p < 0.005) between collaboration scores and compliance improvement which emphasizes the advantages of human-AI cooperation. Thus it is inferred that deep learning not only boosts risk detection productivity but also eliminates human bias at the same time and on top of that, combining the AI and compliance officer has led to more informed, accountable, and ethically sound financial decisions. Therefore, this study has a double contribution to the AI governance discussion in finance by providing not only an empirical but also a theoretical basis for hybrid decision models for AML ordering and optimizing.