iDenfy has updated its hybrid KYC and AML platform, combining automated detection with in-house human review to address rising synthetic identity fraud.
According to the company's in-house research team, fraud rates have risen by at least 15% across several industries, with high-risk sectors such as iGaming among the most affected.
The updated platform retains a multi-layered verification architecture but places renewed emphasis on human review at every stage, not only for flagged cases. When a user submits identity documents, the system cross-references more than 3.000 government-issued document types from over 200 countries and territories. Initial biometric recognition and 3D liveness detection are completed in under three minutes, with deepfake filtering applied at the point of registration before the process advances further.
Where the model diverges from automation-only approaches is in what follows. Rather than routing only suspicious submissions to compliance staff, iDenfy's in-house team manually reviews every identity audit. This includes cross-checking document-to-face similarities and assessing country-level risk signals. Moreover, the company reports that its KYC specialists operate around the clock, reducing the risk of processing delays.
The rationale behind this approach reflects a structural limitation acknowledged in the press release: rule-based and AI-only systems are designed to detect known fraud patterns, while generative AI tools are increasingly used to produce novel attack vectors specifically built to evade those models. Purely manual processes, meanwhile, cannot scale to the volumes required by modern digital onboarding.
AML screening and business verification
The third layer of the platform covers AML screening and ongoing monitoring, including continuous checks against global sanctions lists, Politically Exposed Persons (PEP) databases, and adverse media sources throughout a user's account lifecycle. For regulated business clients, iDenfy has also launched an AI-powered company reviewer that automates Know Your Business (KYB) decisions by cross-referencing company documents and conducting Ultimate Beneficial Owner (UBO) verification against government registry data. Cases with partial matches or flagged signals are routed to the human review team for further assessment.
The updates come against a backdrop of significant financial exposure. According to figures cited by the company, lenders in the US lost more than USD 35 billion in 2025 due to fraudulent identities linked to new bank accounts. Estimates suggest businesses could lose between USD 20 billion and USD 40 billion annually to synthetic identity fraud, attacks that typically combine a stolen Social Security number with fabricated biometric data to bypass verification systems.
Synthetic identity fraud presents a particular challenge during customer onboarding. Fraudsters increasingly combine valid data points, such as real addresses or genuine Social Security numbers, with fabricated names, borrowed birth dates, or AI-generated document images. Deepfake video technology is also being deployed in biometric verification stages to circumvent liveness detection checks.
A company official noted that the organisation has increased investment in its compliance team during the first quarter of 2026, describing skilled personnel as central to detecting fraud patterns and contextual anomalies that automated systems are not equipped to identify.
