US banks are responding to reports that “AI agents” may be gaining unauthorized access to customer accounts by using customers’ login credentials. Multiple outlets describe bank efforts to tighten security and limit the ability of third-party agents to sign in or interact with accounts.
Decrypt says major banks, including Morgan Stanley, JPMorgan, Bank of America, and Citi, are working to restrict third-party AI agents that log in using customers’ credentials. Gizmodo frames the issue around customers effectively “granting” access to bank accounts to AI agents, highlighting concerns that consumers may not realize the access they enable.
Channel NewsAsia reports the issue as US banks detect unauthorized AI agents in customer accounts, citing Axios. This coverage places emphasis on detection by banks and the steps being considered to stop unauthorized activity, though details of any specific restrictions are not provided in the information available here.
