A divide is emerging in banking over AI agents that can access and manage customer finances. JPMorgan Chase, Bank of America and Wells Fargo are restricting or blocking third-party agents, citing security and liability concerns after detecting unauthorized activity. [2, 3, 7, 14]
The banks say they struggle to distinguish bots from customers when users share login credentials, a practice that often violates banks’ terms of service. [2, 3, 7, 14]
Smaller banks, including Lead Bank and Increase Bank, are treating the trend as a growth opportunity. They are partnering with fintech firms to create controlled environments for AI agents, including separate accounts with spending caps and limits on transaction types. These accounts let agents perform defined tasks without access to customers’ primary funds. [1]
Visa and Mastercard are also developing infrastructure to issue unique credentials to agents, with the aim of improving security and tracking. [2, 13]