In a growing concern for financial institutions, the rise of instant payments has led to an irreversible loss of funds. According to recent data from PYMNTS Intelligence, 40% of financial institutions lost more money to fraud last year, with scams accounting for 23% of fraudulent transactions and experiencing a 56% year-over-year rise.
The problem is compounded by the fact that once funds leave an account, they cannot be recalled. This has led banks to invest in AI-powered solutions to investigate and recover stolen funds after the transaction clears. Nasdaq Verafin's Agentic AI Workforce, for example, includes two new role-based agents: the Agentic Fraud Analyst and the Agentic AML Analyst. These agents will automate investigative work currently performed manually by fraud and compliance teams.
In India, the Reserve Bank Innovation Hub has launched MuleHunter.AI, an AI system operational across 26 banks that detects about 20,000 mule accounts per month. Mule accounts are intermediary accounts used to route stolen funds through multiple banks before withdrawing them. The Indian Cyber Crime Coordination Centre reported identifying 2.65 million first-layer mule accounts as of December 31, facilitating the theft of nearly $2.4 billion.
Meanwhile, JPMorgan Chase and ACI Worldwide have announced a partnership to embed JPMorgan's Kinexys Liink account verification directly into ACI Worldwide's enterprise fraud platform. This move aims to apply consistent controls across payment rails before funds leave the account.
The use of AI in recovering stolen funds is becoming increasingly important as traditional methods become impractical due to the speed at which instant payments clear. By reconstructing transaction patterns and connecting related activity across institutions, banks hope to stay ahead of cybercriminals and recover lost funds.