Gartner defines fraud detection in banking payments as platforms that use machine learning (ML) models and business rule engines (BREs) to detect and prevent criminal activities related to money movements that aim to defraud the banks and its customers. Banks use these platforms to determine the risk associated with events, such as payments (including real-time payments), and typically include modules for transaction monitoring, a decision engine and case investigations. A high-risk score may initiate a further review to determine if it is a true positive (fraud) or a false positive (not fraud). Modern platforms that incorporate ML models and BREs are capable of monitoring many account actions and use data from multiple sources. They are generally not used for identity verification, internal fraud, physical controls at branches and ATMs or accounts payable functions.
Gartner defines the OFD market as the market for solutions that detect and prevent fraudulent actions within digital channels (browsers and mobile apps). OFD solutions provide a spectrum of capabilities within digital channels to prevent direct and indirect financial losses and to mitigate risks. Their core capabilities: Mitigate the activity of malicious automated bots; Detect account takeover (ATO) attacks and trigger remedial actions; Detect fraudulent activity in high-risk events along the digital customer journey, such as when customers make payments, transfer funds, perform account management actions or access personally identifiable information (PII). (Retired as of May-06-2026).