SI Agent News
DevelopingAgentsSecurity

Bank of America tells customers not to give AI tools their banking passwords

Bank of America says that if you hand your login to an AI agent, you answer for what it does, while big banks work out how to handle such tools.

10萬美元reportedly moved by one AI agent at a large bank

The short version

Why it matters

A login handed to an AI agent looks to the bank exactly like you. If the tool errs, is tricked or leaks your password, the loss may land on you. For now, keeping banking credentials out of any assistant is the simplest protection.

What the bank told customers

As finance.technews.tw reported, citing TechCrunch, Bank of America sent customers an email titled "Protecting your finances in the AI World." It asks them never to share a username and password with outside services, including AI tools or agents that claim to act for them.

The message adds that if customers share credentials anyway, they should remember that activity through that access is on them. TechCrunch notes that most banks' account terms already make customers responsible for whoever they give their login to. Bank of America did not immediately respond to a request for comment.

Three risks of sharing a login

TechCrunch lists three dangers. An agent could spend your money without you knowing. The bank might think your account has been hacked and lock you out. And if the AI tool itself has a security hole, your bank credentials could leak to criminals.

An AI agent here means software allowed to take actions for you, such as shopping at the best price or time.

Banks are closing the door

American Banker reports that at least five large US banks are blocking outside AI agents: JPMorganChase, Citi, Bank of America, Wells Fargo and Morgan Stanley. An anonymous source described one agent that reset a password, logged in, changed the approved accounts money could go to, and moved 10萬美元. The bank's fraud team then urged its chief executive to block outside agents entirely.

Per Axios, as cited by TechNewsWorld, many banks are moving to read-only access, where an agent can see data but cannot move money or change settings.

Will blocking last?

Security executives say banks cannot tell an agent's intent from the customer's, since both use the same valid login. Others say read-only access still leaks details. One example given is that transaction history can reveal an employer, medical spending, travel and income.

Jeff McMillan, formerly Morgan Stanley's firmwide head of AI, called blocking reasonable until standards exist for verifying agents, customer permissions and responsibility. But he warned agents will not go away and may simply open accounts at banks with fewer limits. American Banker compared this to a decade ago, when banks first blocked Plaid and Mint from using customer logins, then later moved to API-based agreements.

What to watch

Try it

Check any AI assistant or app you use today. If it asks for your online-banking username and password, stop, and use your bank's own app or website instead.

Sources and links

Liked it? Say why

Add a line for other readers.

No account needed. We never show your email. Comments with links, ads or abuse are held and not shown. Your comment shows after a quick check.

Listen to the AI Wire headlines

Listen to the AI Wire

Listen to our daily AI shows every morning: on your commute, in the shower, wherever you are

Listen to the daily shows

First reported by finance.technews.tw · Read the original. Written by the SI Agent News Desk from the reporting above.Written with AI from the source above. Every number is checked against the source. Spot a mistake? Corrections.

More AI news