Newsmax
Artificial intelligence is becoming a security nightmare for major U.S. banks, which have discovered unauthorized AI agents accessing customer accounts, Axios reported.
In some cases, the AI agents appear to act independently of account holder instructions.
Banking giants including JPMorgan Chase, Bank of America, Morgan Stanley, Wells Fargo, and Citigroup are scrambling to protect customers as increasingly sophisticated AI assistants penetrate banking systems faster than institutions can develop safeguards, Axios reported Thursday.
The growing threat has prompted several banks to block third-party AI agents and rewrite their terms of service to prevent potentially unauthorized financial transactions.
According to Axios, banking executives are particularly concerned about AI agents gaining access to customer login credentials and potentially transferring money or making purchases without explicit instructions.
Some institutions have detected agents contacting customer service centers or taking actions that appear to go beyond what account holders authorized.
One banking executive described the situation as horrifying, while others acknowledged that banks struggle to distinguish legitimate customer activity from automated agents operating without proper authorization.
Morgan Stanley has begun blocking third-party agents from its platforms, while JPMorgan, Bank of America, and Citi are implementing similar restrictions.
A Morgan Stanley spokesperson told Axios the company has not observed instances of “rogue AI agents accessing our platforms.”
The concerns extend beyond cybersecurity to the potential disruption of America’s banking system.
The Financial Times reported last week that financial institutions fear AI assistants could automatically transfer billions of dollars from traditional checking accounts into higher-yielding alternatives, threatening a major source of inexpensive funding for banks.
Apollo chief economist Torsten Slok warned that AI agents could shift household deposits earning approximately 0.1% interest into accounts offering returns of 3.3% to 5%.
JPMorgan, Bank of America and Wells Fargo collectively hold $1.6 trillion in non-interest-bearing deposits.
Paying 3% interest on those funds would cost approximately $47 billion annually, according to the Financial Times.
Bank of America CEO Brian Moynihan emphasized the importance of maintaining human oversight of financial transactions.
“We strongly believe there should be a human in the loop to authorise money moving out of your account,” Moynihan told CNBC.
JPMorgan CEO Jamie Dimon similarly warned against allowing unrestricted access by outside AI systems.
“We have to make sure we protect our customers in the right way,” Dimon said, the Financial Times reported.
Banks are exploring compromises that would permit AI assistants to review account information without authorizing transactions.
Meanwhile, financial institutions are developing contingency plans to restore customer access and secure accounts if AI-related security failures occur.
Despite their substantial financial resources and cybersecurity experience, banking executives acknowledge that rapidly advancing AI technology presents risks that traditional security measures may struggle to contain.