Banks Must Unify Identity Systems and Speed Decisions to Combat AI Fraud

Generative AI can now fabricate selfies, voices and identity documents, undermining the remote verification methods banks use for account openings, logins and customer service centers. US Federal Trade Commission data showed fraud losses reached $12.5 billion in 2024, up 25% year on year, making cross-channel identity risk a major concern for financial institutions.

On March 19, 2026, payment processor i2c said deepfake fraud attempts in financial services had risen by more than 2,000% globally. It said banks should integrate device, behavioral and cross-channel signals while embedding real-time scoring and authorization. On April 28, Entersekt further argued that, in millisecond-speed payment environments, banks must verify whether AI agents have user authorization.