FinCEN Warns Financial Institutions on Use of "Deepfake Media" to Commit Fraud

FinCEN alerted financial institutions to the growing use of deepfake media by fraudsters who use AI generated content to fool identity verification processes in schemes targeting the financial sector.

FinCEN warned that the use of deepfake technology—"a type of synthetic content that use [AI/ML] to create realistic but inauthentic videos, pictures, audio, and text"— has risen significantly since 2023. In the Alert, FinCEN identified red flags to help financial institutions detect deepfake media in customer identity documents and institutional vulnerabilities that criminals exploit using AI.

FinCEN highlighted several warning signs associated with deepfake-related fraud, including:

• a customer presenting multiple identity documents that contain conflicting information;

• identity photos or documents that appear altered or do not align with the customer's profile (e.g. a mismatch between photo age and stated date of birth);

• a customer's refusal to engage in multifactor authentication or failure to authenticate identity during video verification due to alleged technical issues;

• high-volume transactions with potentially high-risk payees, such as gambling websites or digital asset exchanges;

• new accounts showing unusual transaction patterns, including rapid withdrawals following deposits;

• geographic or device data inconsistencies in customer profiles; and

• reports of stolen personal information (e.g. Social Security numbers) linked with synthetic identities.

FinCEN urged financial institutions to file suspicious activity reports using the key term "FIN-2024-DEEPFAKEFRAUD" as soon as possible if deepfake-related fraud is suspected. Further, FinCEN advised financial institutions to enhance due diligence and identity verification methods, such as using multifactor authentication, reverse-image searches and live verification.