Detecting Synthetic Identity Fraud Through Graphs, Signals, and Machine Learning

Socure’s analysis describes a synthetic fraud ring that conducted more than 7,000 money-transfer transactions over roughly four months using about 400 fabricated identities, all routed through a single IP address and kept below a $3,000 threshold that exempted them from full KYC checks. The operation exploited the combination of absent upstream identity verification in the affected flow and the transaction-based exemption, despite Socure’s synthetic-fraud detection tools being active in other areas of the institution. The identities showed widespread inconsistencies across names, dates of birth, phone numbers, emails, and addresses, while scripted variations, overnight activity, high transaction velocity, and concentrated payment recipients indicated automation and coordination. Socure’s no-SSN Sigma Synthetic model reportedly identified every evaluated identity, producing a median fraud score of 0.99, with related email and phone risk models reinforcing the finding. Graph analysis also linked shared identity attributes across multiple organizations, revealing broader reuse of fabricated or altered information. The analysis recommends extending identity verification to the uncovered flow and replacing or modifying the fixed sub-$3,000 KYC exemption with risk-based controls, while treating bank-account “penny drop” checks only as a supplementary measure because they verify account access rather than identity.

No tracked trend matches for this post yet.

Use this post, company, and trend context to find content marketing opportunities, perform competitive analysis, or address product feature gaps via the Plushcap MCP server or the Plushcap API.