Payment decisions translate behavioral signals into immediate financial exposure, making payments a distinct and high-stakes risk domain. This chapter covers authorization-time fraud scoring, step-up and policy decisioning, post-authorization monitoring, dispute and chargeback management, and payout controls. It explains why payment ground truth is biased and delayed, why calibration matters when scores drive thresholds, and how latency budgets constrain model complexity. The chapter also addresses broader financial exposure, including credit-like risk in instalments or platform financing, and the compliance context that shapes data usage and controls. Effective systems balance approval rates with loss reduction by combining real-time scoring with rules, segmentation, and structured feedback loops. Monitoring links decisions to measurable outcomes such as chargeback rates, fraud loss, operational load, and customer friction. The chapter also explains how risk signals interact with payment method choice and regional constraints, and why combining model scores with policy rules is essential for controllable outcomes. By grounding modeling in the payment lifecycle, the chapter clarifies which decisions belong at authorization versus downstream recovery stages. By treating payment risk as an end-to-end decision system rather than a single classifier, the chapter provides a practical framework for safeguarding financial trust while preserving legitimate transaction flow.