September 2022 Summaries

Financial crime has expanded beyond application fraud and account takeover, with cryptocurrencies, pandemic relief programs, social media, and stolen personal data creating new opportunities for money laundering, identity theft, synthetic identities, and fraudulent claims. Examples include the alleged laundering of Bitcoin stolen from Bitfinex through transaction-obscuring techniques and synthetic identities, alleged misuse of COVID-19 relief funds by opportunistic “citizen” fraudsters, and street gangs shifting from drug activity toward large-scale identity theft and unemployment or credit-card fraud. The account argues that online resources and social media have helped novice fraudsters learn techniques, connect with more experienced actors, and sometimes expose themselves by displaying illicit spending. It cites forecasts of growing identity-theft losses and describes fraud rings as increasingly organized and geographically broad, using data from sources such as stolen mail, public records, dark-web markets, and online listings. The piece concludes that organizations need adaptive identity-verification and fraud-prevention systems, promoting Socure’s machine-learning platform as a tool designed to identify evolving fraud methods.

Socure argues that a sudden increase in new-applicant declines may indicate successful detection of a large fraud attack rather than a deterioration in customer conversion, particularly when application volume rises while approved-user volume remains stable. It recommends monitoring approval volumes and fraud-score distributions instead of approval rates alone, since attacks may generate many high-risk applications and suppress acceptance percentages. In a cited case, a financial institution experienced a 20% acceptance-rate decline during an early-morning fraud campaign that repeatedly changed email addresses and phone numbers; Sigma Identity Fraud reportedly identified more than 80% of attempts at an acceptable false-positive rate, and revised decision rules raised capture to 93% with a 1% increase in reviews. The company attributes its detection capabilities to machine learning, broad identity analysis across personal and behavioral data, consortium feedback data, and adjustable controls using email, phone, and address risk and correlation scores, which it says can help institutions adapt quickly while continuing to approve legitimate applicants.

Immigration could significantly influence US economic growth, with FWD.us projecting that high legal immigration could contribute to roughly $50 trillion in GDP by 2050, compared with about $37 trillion under low immigration, while new-to-country residents already represent nearly 14% of the population. However, many immigrants face barriers to financial and retail onboarding because they may lack US credit histories or conventional identification records, and legacy identity-verification systems can struggle with international naming conventions and rely on methods that may create disproportionate friction for certain demographic groups. The passage argues that biased data, machine-learning models, and outdated entity-resolution techniques can limit access to credit and other services for new-to-country consumers. It presents Socure’s data-driven identity verification, entity resolution, clustering, and machine-learning tools as an alternative, claiming improved identification pass rates for Hispanic and Asian consumers and suggesting that more inclusive onboarding can expand access while helping organizations reach growing customer segments.