SEON has introduced an expanded set of artificial intelligence features designed to help fraud prevention and anti-money laundering (AML) teams move from raw data to immediate action while improving transparency. The new suite aims to cut manual review time by up to 50% and reduce the need for analysts to switch between platforms.
Improving fraud investigations with explainable AI
The updated platform automatically detects connections between users, highlights risk signals using colour-coded indicators, and introduces an intelligent AML screening agent. Unlike black-box systems, SEON’s AI is built to be explainable. Analysts can see exactly how risk scores are calculated and which data points influenced each decision.
Tamas Kadar, co-founder and chief executive officer at SEON, said the company’s first-party data approach helps analysts make confident decisions. “Fraud teams don’t only need more data; they need better context. By capturing risk signals at the earliest customer touchpoints, our AI turns massive data volumes into clear, actionable intelligence,” he said.
This approach reflects wider industry trends. SEON’s 2025 Digital Fraud Outlook found that 76% of businesses are increasing AI investments to support rather than replace human judgment. By explaining each risk score, SEON aims to improve trust and accelerate responses.
New features for faster, clearer analysis
SEON’s AI-driven enhancements cover every stage of fraud and AML investigations. Risk signals now surface activity levels across data points such as email, phone, device, operating system, and IP address, allowing analysts to spot critical triggers instantly.
The new similarity ranking feature connects and prioritises linked users based on shared devices, behaviours, IPs, and contacts, eliminating the need for manual graph-building. AI investigation summaries turn complex digital footprints into clear narratives, showing why specific activities were flagged.
For greater adaptability, the natural language rule and filter builder allows analysts to describe detection logic in plain English, with AI automatically creating complex rules. The AML screening agent also helps by identifying false positives and prioritising the most relevant alerts.
Kimon Chalkias, risk control team leader at Kaizen Gaming, said the new tool has improved agility. “Since implementing SEON’s natural language rule and filter builder, we can create sophisticated detection rules in minutes by simply describing what we want in plain English. It’s eliminating the lengthy testing cycles and technical back-and-forth we used to experience, and we expect it will help us adapt to new fraud patterns much faster,” he said.
Data-driven precision for modern compliance needs
SEON’s AI engine operates on over 900 real-time, first-party data signals, including behavioural and digital footprint data captured as it happens. This dynamic approach is designed to increase accuracy compared to systems relying on static, third-party sources.
Industry experts say such solutions are crucial as financial crime grows more sophisticated. Chuck Subrt, fraud and AML practice director at Datos Insights, said: “Numerous industries such as financial technology face a critical inflection point where traditional investigation methods may no longer keep pace with sophisticated financial crime. With mounting regulatory pressures, resource constraints, and an increasingly complex threat environment, investigation optimisation has evolved from operational improvement to a strategic necessity.”
By integrating advanced AI with explainable outputs, SEON aims to make fraud and AML investigations faster and more reliable, helping organisations meet compliance standards while improving analyst productivity.
