Big Banks Scramble to Block Unauthorized AI Agents From Customer Accounts, Axios Reports

Major U.S. banks are finding third-party AI agents operating inside customer accounts and are rushing to contain it, according to a Breitbart News report that draws on Axios. Fox News aired a segment on the same topic Monday morning.

What banks are seeing. Axios reports that Morgan Stanley, Bank of America, JPMorgan, Wells Fargo and others are discovering AI agents that in some cases appear to take actions the account holder never asked for. One bank official called it "horrifying." Another cited "grave concerns about agents gaining access to customer accounts and taking steps that could impact user finances."

How it happens. A customer hands login details to an AI assistant to manage part of their finances. That already breaks most banks' rules, which require a human to log in. The larger risk comes when the user has also allowed the agent to buy things or move money. Banks say agents have contacted call centers and acted without direct user instructions, and that they struggle to tell whether an agent was authorized or has "gone rogue."

The response. Morgan Stanley has started blocking third-party agents outright. JPMorgan, Bank of America and Citi are restricting access. Many banks are settling on read-only access, similar to what budgeting apps get. They are also running "day after" planning to lock down or restore access fast if an agent causes harm. A Chase spokesperson said "customers should not give their usernames and passwords to anyone."

The pushback. A Morgan Stanley spokesperson denied reports of "rogue AI agents accessing our platforms."

Note for readers: this is a secondhand account of Axios's reporting, written Oct. 22, so it is older than most items we run today. We did not read the Axios piece, and the bank officials are unnamed. Breitbart's closing paragraph promotes a book by one of its own editors.